Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 1. Introduction

1.1 Historical Perspective and Materials Science

1.1.1 Historical Perspective

Materials are so important in the development of human civilization that the historians
have identified early periods of civilization by the name of most significantly used
material, e.g.: Stone Age, Bronze Age. This is just an observation made to showcase the
importance of materials and their impact on human civilization. It is obvious that
materials have affected and controlling a broad range of human activities through
thousands of decades.

From the historical point of view, it can be said that human civilization started with Stone
Age where people used only natural materials, like stone, clay, skin, and wood for the
purposes like to make weapons, instruments, shelter, etc. Thus the sites of deposits for
better quality stones became early colonies of human civilization. However, the
increasing need for better quality tools brought forth exploration that led to Bronze Age,
followed by Iron Age. When people found copper and how to make it harder by alloying,
the Bronze Age started about 3000 BC. The use of iron and steel, a stronger material that
gave advantage in wars started at about 1200 BC. Iron was abundant and thus availability
is not limited to the affluent. This commonness of the material affected every person in
many aspects, gaining the name democratic material. The next big step in human
civilization was the discovery of a cheap process to make steel around 1850 AD, which
enabled the railroads and the building of the modern infrastructure of the industrial world.
One of the most significant features of the democratic material is that number of users
just exploded. Thus there has been a need for human and material resources for centuries,
which still going strong. Its being said and agreed that we are presently in Space Age
marked by many technological developments towards development materials resulting in
stronger and light materials like composites, electronic materials like semiconductors,
materials for space voyage like high temperature ceramics, biomaterials, etc.

In summary, materials constitute foundation of technology. The history of human

civilization evolved from the Stone Age to the Bronze Age, the Iron Age, the Steel Age,
and to the Space Age (contemporaneous with the Electronic Age). Each age is marked by
the advent of certain materials. The Iron Age brought tools and utensils. The Steel Age
brought railroads, instruments, and the Industrial Revolution. The Space Age brought the
materials for stronger and light structures (e.g., composite materials). The Electronic Age
brought semiconductors, and thus many varieties of electronic gadgets.

1.1.2 Materials Science

As engineering materials constitute foundation of technology, its not only necessary but
a must to understand how materials behave like they do and why they differ in properties.
This is only possible with the atomistic understanding allowed by quantum mechanics
that first explained atoms and then solids starting in the 1930s. The combination of
physics, chemistry, and the focus on the relationship between the properties of a material
and its microstructure is the domain of Materials Science. The development of this
science allowed designing materials and provided a knowledge base for the engineering
applications (Materials Engineering).

Important components of the subject Materials Science are structure, properties,

processing, and performance. A schematic interrelation between these four components is
shown in figure 1.1.

Structure

Performance

Properties Processing

Figure 1.1: Interrelation between four components of Materials Science.

1.2 Why Study Materials Science and Engineering? and Classification of Materials?

1.2.1 Why Study Materials Science and Engineering?

All engineers need to know about materials. Even the most "immaterial", like software or
system engineering depend on the development of new materials, which in turn alter the
economics, like software-hardware trade-offs. Increasing applications of system
engineering are in materials manufacturing (industrial engineering) and complex
environmental systems.

Innovation in engineering often means the clever use of a new material for a specific
application. For example: plastic containers in place of age-old metallic containers. It is
well learnt lesion that engineering disasters are frequently caused by the misuse of
materials. So it is vital that the professional engineer should know how to select materials
which best fit the demands of the design - economic and aesthetic demands, as well as
demands of strength and durability. Beforehand the designer must understand the
properties of materials, and their limitations. Thus it is very important that every engineer
must study and understand the concepts of Materials Science and Engineering. This
enables the engineer

To select a material for a given use based on considerations of cost and

performance.
To understand the limits of materials and the change of their properties with use.
To be able to create a new material that will have some desirable properties.
To be able to use the material for different application.

1.2.2 Classification of Materials

Like many other things, materials are classified in groups, so that our brain can handle
the complexity. One can classify them based on many criteria, for example crystal
structure (arrangement of atoms and bonds between them), or properties, or use. Metals,
Ceramics, Polymers, Composites, Semiconductors, and Biomaterials constitute the main
classes of present engineering materials.

Metals: These materials are characterized by high thermal and electrical conductivity;
strong yet deformable under applied mechanical loads; opaque to light (shiny if
polished). These characteristics are due to valence electrons that are detached from
atoms, and spread in an electron sea that glues the ions together, i.e. atoms are bound
together by metallic bonds and weaker van der Waalls forces. Pure metals are not good
enough for many applications, especially structural applications. Thus metals are used in
alloy form i.e. a metal mixed with another metal to improve the desired qualities. E.g.:
aluminum, steel, brass, gold.

Ceramics: These are inorganic compounds, and usually made either of oxides, carbides,
nitrides, or silicates of metals. Ceramics are typically partly crystalline and partly
amorphous. Atoms (ions often) in ceramic materials behave mostly like either positive or
negative ions, and are bound by very strong Coulomb forces between them. These
materials are characterized by very high strength under compression, low ductility;
usually insulators to heat and electricity. Examples: glass, porcelain, many minerals.
Polymers: Polymers in the form of thermo-plastics (nylon, polyethylene, polyvinyl
chloride, rubber, etc.) consist of molecules that have covalent bonding within each
molecule and van der Waals forces between them. Polymers in the form of thermo-sets
(e.g., epoxy, phenolics, etc.) consist of a network of covalent bonds. They are based on
H, C and other non-metallic elements. Polymers are amorphous, except for a minority of
thermoplastics. Due to the kind of bonding, polymers are typically electrical and thermal
insulators. However, conducting polymers can be obtained by doping, and conducting
polymer-matrix composites can be obtained by the use of conducting fillers. They
decompose at moderate temperatures (100 400 C), and are lightweight. Other properties
vary greatly.

Composite materials: Composite materials are multiphase materials obtained by

artificial combination of different materials to attain properties that the individual
components cannot attain. An example is a lightweight brake disc obtained by embedding
SiC particles in Al-alloy matrix. Another example is reinforced cement concrete, a
structural composite obtained by combining cement (the matrix, i.e., the binder, obtained
by a reaction known as hydration, between cement and water), sand (fine aggregate),
gravel (coarse aggregate), and, thick steel fibers. However, there are some natural
composites available in nature, for example wood. In general, composites are classified
according to their matrix materials. The main classes of composites are metal-matrix,
polymer-matrix, and ceramic-matrix.

Semiconductors: Semiconductors are covalent in nature. Their atomic structure is

characterized by the highest occupied energy band (the valence band, where the valence
electrons reside energetically) full such that the energy gap between the top of the
valence band and the bottom of the empty energy band (the conduction band) is small
enough for some fraction of the valence electrons to be excited from the valence band to
the conduction band by thermal, optical, or other forms of energy. Their electrical
properties depend extremely strongly on minute proportions of contaminants. They are
usually doped in order to enhance electrical conductivity. They are used in the form of
single crystals without dislocations because grain boundaries and dislocations would
degrade electrical behavior. They are opaque to visible light but transparent to the
infrared. Examples: silicon (Si), germanium (Ge), and gallium arsenide (GaAs, a
compound semiconductor).

Biomaterials: These are any type material that can be used for replacement of damaged
or diseased human body parts. Primary requirement of these materials is that they must
be biocompatible with body tissues, and must not produce toxic substances. Other
important material factors are: ability to support forces; low friction, wear, density, and
cost; reproducibility. Typical applications involve heart valves, hip joints, dental
implants, intraocular lenses. Examples: Stainless steel, Co-28Cr-6Mo, Ti-6Al-4V, ultra
high molecular weight poly-ethelene, high purity dense Al-oxide, etc.

1.3 Advanced Materials, Future Materials, and Modern Materials needs

1.3.1 Advanced Materials

These are materials used in High-Tech devices those operate based on relatively intricate
and sophisticated principles (e.g. computers, air/space-crafts, electronic gadgets, etc.).
These materials are either traditional materials with enhanced properties or newly
developed materials with high-performance capabilities. Hence these are relatively
expensive. Typical applications: integrated circuits, lasers, LCDs, fiber optics, thermal
protection for space shuttle, etc. Examples: Metallic foams, inter-metallic compounds,
multi-component alloys, magnetic alloys, special ceramics and high temperature
materials, etc.

1.3.2 Future Materials

Group of new and state-of-the-art materials now being developed, and expected to have
significant influence on present-day technologies, especially in the fields of medicine,
manufacturing and defense. Smart/Intelligent material system consists some type of
sensor (detects an input) and an actuator (performs responsive and adaptive function).
Actuators may be called upon to change shape, position, natural frequency, mechanical
characteristics in response to changes in temperature, electric/magnetic fields, moisture,
pH, etc.

Four types of materials used as actuators: Shape memory alloys, Piezo-electric ceramics,
Magnetostrictive materials, Electro-/Magneto-rheological fluids. Materials / Devices used
as sensors: Optical fibers, Piezo-electric materials, Micro-electro-mechanical systems
(MEMS), etc.

Typical applications: By incorporating sensors, actuators and chip processors into system,
researchers are able to stimulate biological human-like behavior; Fibers for bridges,
buildings, and wood utility poles; They also help in fast moving and accurate robot parts,
high speed helicopter rotor blades; Actuators that control chatter in precision machine
tools; Small microelectronic circuits in machines ranging from computers to
photolithography prints; Health monitoring detecting the success or failure of a product.

1.3.3 Modern Materials needs

Though there has been tremendous progress over the decades in the field of materials
science and engineering, innovation of new technologies, and need for better
performances of existing technologies demands much more from the materials field.
More over it is evident that new materials/technologies are needed to be environmental
friendly. Some typical needs, thus, of modern materials needs are listed in the following:

Engine efficiency increases at high temperatures: requires high temperature

structural materials
 Use of nuclear energy requires solving problem with residues, or advances in
nuclear waste processing.
Hypersonic flight requires materials that are light, strong and resist high
temperatures.
Optical communications require optical fibers that absorb light negligibly.
Civil construction materials for unbreakable windows.
Structures: materials that are strong like metals and resist corrosion like plastics.

References

1. M. F. Ashby and D. R. H. Jones, Engineering Materials 1, An introduction to

Their Properties and Applications, second edition, Butterworth-Heinemann,
Woburn, UK, 1996
2. William D. Callister, Jr, Materials Science and Engineering An introduction,
sixth edition, John Wiley & Sons, Inc. 2004.
3. V. Raghavan, Materials Science and Engineering, third edition, Prentice Hall of
India Private Limited, New Delhi, 1990.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 2. Atomic Structure, Interatomic Bonding and Structure

of Crystalline Solids

2.1 Atomic Structure and Atomic Bonding in Solids

2.1.1 Atomic Structure

Atoms are composed of electrons, protons, and neutrons. Electrons and protons are
negative and positive charged particles respectively. The magnitude of each charged
particle in an atom is 1.6 10-19 Coulombs.

The mass of the electron is negligible with respect to those of the proton and the neutron,
which form the nucleus of the atom. The unit of mass is an atomic mass unit (amu) =
1.66 10-27 kg, and equals 1/12 the mass of a carbon atom. The Carbon nucleus has Z=6,
and A=6, where Z is the number of protons, and A the number of neutrons. Neutrons and
protons have very similar masses, roughly equal to 1 amu each. A neutral atom has the
same number of electrons and protons, Z.

A mol is the amount of matter that has a mass in grams equal to the atomic mass in amu
of the atoms. Thus, a mole of carbon has a mass of 12 grams. The number of atoms in a
mole is called the Avogadro number, Nav = 6.023 1023. Note that Nav = 1 gram/1 amu.

Calculating n, the number of atoms per cm3 of a material of density (g/cm3):

n = N av
M

where M is the atomic mass in amu (grams per mol). Thus, for graphite (carbon) with a
density = 1.8 g/cm3, M =12, we get 6 1023 atoms/mol 1.8 g/cm3 / 12 g/mol) = 9
1022 C atoms/cm3.
For a molecular solid like ice, one uses the molecular mass, M(H2O) = 18. With a density
of 1 g/cm3, one obtains n = 3.3 1022 H2O molecules/cm3. Note that since the water
molecule contains 3 atoms, this is equivalent to 9.9 1022 atoms/cm3.

Most solids have atomic densities around 6 1022 atoms/cm3. The cube root of that
number gives the number of atoms per centimeter, about 39 million. The mean distance
between atoms is the inverse of that, or 0.25 nm. This is an important number that gives
the scale of atomic structures in solids.

2.1.2 Atomic bonding in solids

In order to understand the why materials behave like they do and why they differ in
properties, it is necessary that one should look at atomic level. The study primarily
concentrates on two issues: what made the atoms to cluster together, and how atoms are
arranged. As mentioned in earlier chapter, atoms are bound to each other by number of
bonds. These inter-atomic bonds are primarily of two kinds: Primary bonds and
Secondary bonds. Ionic, Covalent and Metallic bonds are relatively very strong, and
grouped as primary bonds, whereas van der Waals and hydrogen bonds are relatively
weak, and termed as secondary bonds. Metals and Ceramics are entirely held together by
primary bonds - the ionic and covalent bonds in ceramics, and the metallic and covalent
bonds in metals. Although much weaker than primary bonds, secondary bonds are still
very important. They provide the links between polymer molecules in polyethylene (and
other polymers) which make them solids. Without them, water would boil at -80C, and
life as we know it on earth would not exist.

Ionic Bonding: This bond exists between two atoms when one of the atoms is negative
(has an extra electron) and another is positive (has lost an electron). Then there is a
strong, direct Coulomb attraction. Basically ionic bonds are non-directional in nature. An
example is NaCl. In the molecule, there are more electrons around Cl, forming Cl- and
fewer electrons around Na, forming Na+. Ionic bonds are the strongest bonds. In real
solids, ionic bonding is usually exists along with covalent bonding.

Na Cl

Fig.1 Schematic representation of ioning bonding. Here, Na is giving an electron to Cl to

have stable structure
Covalent Bonding: In covalent bonding, electrons are shared between the atoms, to
saturate the valency. The simplest example is the H2 molecule, where the electrons spend
more time in between the nuclei of two atoms than outside, thus producing bonding.
Covalent bonds are stereo-specific i.e. each bond is between a specific pair of atoms,
which share a pair of electrons (of opposite magnetic spins). Typically, covalent bonds
are very strong, and directional in nature. The hardness of diamond is a result of the fact
that each carbon atom is covalently bonded with four neighboring atoms, and each
neighbor is bonded with an equal number of atoms to form a rigid three-dimensional
structure.

H H

Figure 2. Schematic representation of covalent bond in Hydrogen molecule (sharing of

electrons)

Metallic Bonding: Metals are characterized by high thermal and electrical

conductivities. Thus, neither covalent nor ionic bondings are realized because both types
of bonding localize the valence electrons and preclude conduction. However, strong
bonding does occur in metals. The valence electrons of metals also are delocalized. Thus
metallic bonding can be viewed as metal containing a periodic structure of positive ions
surrounded by a sea of delocalized electrons. The attraction between the two provides the
bond, which is non-directional.
Electron cloud from
the valence electrons
M M M M

M M M M Core

M M M M

M M M M
 Figure 3. Metallic bonding

Fluctuating Induced Dipole Bonds: Since the electrons may be on one side of the atom
or the other, a dipole is formed: the + nucleus at the center, and the electron outside.
Since the electron moves, the dipole fluctuates. This fluctuation in atom A produces a
fluctuating electric field that is felt by the electrons of an adjacent atom, B. Atom B then
polarizes so that its outer electrons are on the side of the atom closest to the + side (or
opposite to the side) of the dipole in A.

Polar Molecule-Induced Dipole Bonds: Another type of secondary bond exists with
asymmetric molecules, also called polar molecules because of positively and negatively
charged regions. A permanent dipole moment arises from net positive and negative
charges that are respectively associated with the hydrogen and chlorine ends of the HCl
molecule, leading to bonding. The magnitude of this bond will be greater than for
fluctuating induced dipoles.

Figure 4. Dipole bond in water

These two kinds of bonds are also called van der Waals bonds. Third type of secondary
bond is the hydrogen bond. It is categorized separately because it produces the strongest
forces of attraction in this category.

Permanent Dipole Bonds / Hydrogen bonding: It occurs between molecules as

covalently bonded hydrogen atoms for example C-H, O-H, F-H share single electron
with other atom essentially resulting in positively charged proton that is not shielded any
electrons. This highly positively charged end of the molecule is capable of strong
attractive force with the negative end of an adjacent molecule. The properties of water are
influenced significantly by the hydrogen bonds/bridges. The bridges are of sufficient
strength, and as a consequence water has the highest melting point of any molecule of its
size. Likewise, its heat of vaporization is very high.
2.2 Crystal Structures, Crystalline and Non-Crystalline materials

2.2.1 Crystal structures

All metals, a major fraction of ceramics, and certain polymers acquire crystalline form
when solidify, i.e. in solid state atoms self-organize to form crystals. Crystals possess a
long-range order of atomic arrangement through repeated periodicity at regular intervals
in three dimensions of space. When the solid is not crystalline, it is called amorphous.
Examples of crystalline solids are metals, diamond and other precious stones, ice,
graphite. Examples of amorphous solids are glass, amorphous carbon (a-C), amorphous
Si, most plastics.

There is very large number of different crystal structures all having long-range atomic
order; these vary from relatively simple structures for metals to exceedingly complex
structures for ceramics and some polymers. To discuss crystalline structures it is useful to
consider atoms as being hard spheres, with well-defined radii. In this scheme, the shortest
distance between two like atoms is one diameter. In this context, use of terms lattice and
unit cell will be handy. Lattice is used to represent a three-dimensional periodic array of
points coinciding with atom positions. Unit cell is smallest repeatable entity that can be
used to completely represent a crystal structure. Thus it can be considered that a unit cell
is the building block of the crystal structure and defines the crystal structure by virtue of
its geometry and the atom positions within.

Important properties of the unit cells are

The type of atoms and their radii R.

Cell dimensions (Lattice spacing a, b and c) in terms of R and
Angle between the axis , ,
a*, b*, c* - lattice distances in reciprocal lattice , *, *, * - angle in reciprocal
lattice
n, number of atoms per unit cell. For an atom that is shared with m adjacent unit
cells, we only count a fraction of the atom, 1/m.
CN, the coordination number, which is the number of closest neighbors to which
an atom is bonded.
APF, the atomic packing factor, which is the fraction of the volume of the cell
actually occupied by the hard spheres. APF = Sum of atomic volumes/Volume of
cell.

Some very common crystal structures and relevant properties are listed in table 2.1.

Table 2.1: Common crystal structures and their properties.

Unit Cell n CN a/R APF

Simple Cubic 1 6 4/4 0.52

 Body-Centered Cubic 2 8 4/ 3 0.68

Face-Centered Cubic 4 12 4/ 2 0.74

Hexagonal Close Packed 6 12 0.74

Figure 2.1: Common metallic crystal structures.

2.2.2 Crystalline and Non-crystalline materials

Single Crystals: Crystals can be single crystals where the whole solid is one crystal.
Then it has a regular geometric structure with flat faces.

Polycrystalline Materials: A solid can be composed of many crystalline grains, not

aligned with each other. It is called polycrystalline. The grains can be more or less
aligned with respect to each other. Where they meet is called a grain boundary.

Non-Crystalline Solids: In amorphous solids, there is no long-range order. But

amorphous does not mean random, since the distance between atoms cannot be smaller
than the size of the hard spheres. Also, in many cases there is some form of short-range
order. For instance, the tetragonal order of crystalline SiO2 (quartz) is still apparent in
amorphous SiO2 (silica glass).

2.3 Miller Indices, Anisotropy, and Elastic behavior of composites

2.3.1 Miller indices:

It is understood that properties of materials depend on their crystal structure, and many of
these properties are directional in nature. For example: elastic modulus of BCC iron is
greater parallel to the body diagonal than it is to the cube edge. Thus it is necessary to
characterize the crystal to identify specific directions and planes. Specific methods are
employed to define crystal directions and crystal planes.

Methodology to define crystallographic directions in cubic crystal:

 - a vector of convenient length is placed parallel to the required direction.
- the length of the vector projection on each of three axes are measured in unit cell
dimensions.
- these three numbers are made to smallest integer values, known as indices, by
multiplying or dividing by a common factor.
- the three indices are enclosed in square brackets, [uvw]. A family of directions is
represented by <uvw>.

Methodology to define crystallographic planes in cubic crystal:

- determine the intercepts of the plane along the crystallographic axes, in terms of
unit cell dimensions. If plane is passing through origin, there is a need to construct
a plane parallel to original plane.
- take the reciprocals of these intercept numbers.
- clear fractions.
- reduce to set of smallest integers.
- The three indices are enclosed in parenthesis, (hkl). A family of planes is
represented by {hkl}.

For example, if the x-, y-, and z- intercepts of a plane are 2, 1, and 3. The Miller indices
are calculated as:
- take reciprocals: 1/2, 1/1, 1/3.
- clear fractions (multiply by 6): 3, 6, 2.
- reduce to lowest terms (already there). => Miller indices of the plane are (362).

Figure 2.2 depicts Miller indices for number of directions and planes in a cubic crystal.

Figure 2.2: Miller indices in a cubic crystal.

Some useful conventions of Miller notation:

- If a plane is parallel to an axis, its intercept is at infinity and its Miller index will
be zero.
 - If a plane has negative intercept, the negative number is denoted by a bar above
the number. Never alter negative numbers. For example, do not divide -1, -1, -1
by -1 to get 1,1,1. This implies symmetry that the crystal may not have!
- The crystal directions of a family are not necessarily parallel to each other.
Similarly, not all planes of a family are parallel to each other.
- By changing signs of all indices of a direction, we obtain opposite direction.
Similarly, by changing all signs of a plane, a plane at same distance in other side
of the origin can be obtained.
- Multiplying or dividing a Miller index by constant has no effect on the orientation
of the plane.
- The smaller the Miller index, more nearly parallel the plane to that axis, and vice
versa.
- When the integers used in the Miller indices contain more than one digit, the
indices must be separated by commas. E.g.: (3,10,13)
- By changing the signs of all the indices of (a) a direction, we obtain opposite
direction, and (b) a plane, we obtain a plane located at the same distance on the
other side of the origin.

More conventions applicable to cubic crystals only:

- [uvw] is normal to (hkl) if u = h, v = k, and w = l. E.g.: (111) [111].

- Inter-planar distance between family of planes {hkl} is given by:

a
d {hkl} =
h + k2 + l2
2

- [uvw] is parallel to (hkl) if hu + kv + lw = 0.

- Two planes (h1k1l1) and (h2k2l2) are normal if h1h2 + k1k2 + l1l2=0.
- Two directions (u1v1w1) and (u2v2w2) are normal if u1u2 + v1v2 + w1w2=0
- Angle between two planes is given by:

h1 h2 + k1 k 2 + l1l 2
cos =
h12 + k12 + l12 h22 + k 22 + l 22

The same equation applies for two directions.

Why Miller indices are calculated in that way?

- Using reciprocals spares us the complication of infinite intercepts.
- Formulas involving Miller indices are very similar to related formulas from
analytical geometry.
- Specifying dimensions in unit cell terms means that the same label can be applied
to any plane with a similar stacking pattern, regardless of the crystal class of the
crystal. Plane (111) always steps the same way regardless of crystal system.

2.3.2 Miller-Bravis indices

Though Miller indices can describe all possible planes through any crystal, Miller-Bravis
indices are used in hexagonal crystal systems. This is because they reveal hexagonal
symmetry more clearly. Although partially redundant, they are used exclusively for
hexagonal systems.

Direction indices are obtained as above where first three indices are representative of
projections of the direction over three co-planar axes in the plane called basal plane while
the last index denotes the projection over the axis perpendicular to the basal plane.
Miller-Bravis indices for a plane are denoted as [uvtw], where t = -(u+v)

In the same procedure, planes in a hexagonal crystal are denoted by (hkil), where i = -
(h+k).

2.3.3 Anisotropy

Its been agreed that many of the materials properties depend on the crystal structure.
However, crystals are not symmetric in all directions, or not the crystal planes same with
respect to atomic density/packing. Different directions in the crystal have different
packing. For instance, atoms along the edge of FCC crystals are more separated than
along its face diagonal. This causes properties to be different in different directions. This
directionality of properties is termed as Anisotropy.

Substances in which measured properties are independent of direction in which they are
measured are called isotropic. Though, in polycrystalline materials, the crystallographic
orientations of individual grains are random, specimen may behave isotropically.

2.3.4 Elastic Behavior of Composites

The idea is that by combining two or more distinct materials one can engineer a new
material with the desired combination of properties (e.g., light, strong, corrosion
resistant). The idea that a better combination of properties can be achieved is called the
principle of combined action.

For example, Pearlitic steel that combines hard and brittle Cementite with soft and ductile
ferrite to get a superior material.

A composite is defined as a artificially made muti-phase material, where constituent

phases are chemically dissimilar and separated by a distinct interface. Thus, composites
shall have at least two or more constituent phases. Many of the composites are made of
two phases one is termed as matrix, which is continuous and surrounds the other phase,
called dispersed phase. Dispersed phase can exist in many forms like particulates, short-
/long- fibers.

Properties of composites depend on

 - Properties of the constituent phases.
- Geometry of dispersed phase (particle size, size distribution, orientation).
- Amount of each constituent phase.

Classification of composites based on geometry of dispersed phase:

- Particle-reinforced (large-particle and dispersion-strengthened)
- Fiber-reinforced (continuous (aligned) and short fibers (aligned or random)
- Structural (laminates and sandwich panels)

Classification of composites based on matrix phase:

- Metal Matrix Composites.
- Polymer Matrix Composites.
- Ceramic Matrix Composites.

Composite properties can be calculated using rule of mixtures. For example, elastic
modulus of a particle reinforced composite bound by the limits given by

E c (u ) = E mVm + E pV p

Em E p
E c (l ) =
E mV p + E p V m

where E and V denote the elastic modulus and volume fraction respectively; c, m, and p
represent composite, matrix, and particulate phases.

However, in case of continuous fiber reinforced composites upper bound is applicable to

condition of longitudinal loading, while the lower bound is applicable to transverse
loading condition. The above equations can be simplified using the following relations:

Vm + V p = 1

When fiber orientation is random, or short and discontinuous fibers are used, rule of
mixtures will be modified as follows to take care of the randomness of fibers:

E cd = E mVm + KE f V f

where K is fiber efficiency parameter that depends on Vf and Ef/Em ratio. Ks values are
less than unity, usually attains a value in the range of 0.1 to 0.6.

Many applications, like in aircraft parts, there is a need for high strength per unit weight
(specific strength). This can be achieved by composites consisting of a low-density (and
soft) matrix reinforced with stiff fibers.

2.4 Structure and properties of polymers

Polymers are common in nature, in the form of wood, rubber, cotton, leather, wood, silk,
proteins, enzymes, starches, cellulose. Artificial polymers are made mostly from oil.
Their use has grown exponentially, especially after WW2 (World War-2). The key factor
is the very low production cost and useful properties (e.g., combination of transparency
and flexibility, long elongation, etc.).

Most polymers are organic, and formed from hydrocarbon molecules. These molecules
can have single, double, or triple carbon bonds. A saturated hydrocarbon is one where
all bonds are single, i.e. the number of atoms is maximum (or saturated). Among this type
are the paraffin compounds, CnH2n+2. In contrast, non-saturated hydrocarbons contain
some double and triple bonds.

Isomers are molecules that contain the same molecules but in a different arrangement.
An example is butane and iso-butane. Some physical properties of hydrocarbons depend
on the isomeric state.

2.4.1 Polymer molecules

Polymer molecules are huge, macromolecules that have internal covalent bonds. For most
polymers, these molecules form very long chains. The backbone is a string of carbon
atoms, often single bonded. Polymers are composed of basic structures called mer
units. A molecule with just one mer is a monomer. Within each molecule / mer atoms are
bonding together by strong covalent bonds. When many mers are together, they form
polymer. Bi-functional monomers may bond with two other units in forming 2-D chain-
like structures; while Tri-functional monomers can form three active bonds, and thus 3-D
molecular network. Examples of polymers are polyvinyl chloride (PVC), poly-tetra-
fluoro-ethylene (PTFE or Teflon), polypropylene, nylon and polystyrene. When all the
mers are the same, the molecule is called a homopolymer. When there is more than one
type of mer present, the molecule is a copolymer.

The mass of a polymer is not fixed, but is distributed around a mean value, since not all
polymer chains will grow same extent. The average molecular weight can be obtained by
averaging the masses with the fraction of times they appear (number-average) or with the
weight fraction of the molecules (weight-average). Another representation of average
chain size is degree of polymerization (n) average number of mer units in a chain. It is
obtained by dividing the average mass of the polymer by the mass of a mer unit.
Numbers of polymer characteristics are affected by the magnitude of the molecular
weight. Short chain polymers usually exist in form of gases or liquids at room
temperature; where as medium range polymers are waxy solids and soft resins. Solid
polymers are commonly having weights ranging between 10K and several million g/mol.

2.4.2 Polymer structures

Polymers consist of large number of molecular chains which are usually not linear;
bending and rotations can occur around single C-C bonds (double and triple bonds are
very rigid). Random kinks and coils in chains along with bending of chains lead to
intertwining and entanglement of neighboring chains, situation like in the spaghetti
structure. These characteristic entanglements are responsible for a number of properties
specific to polymers, e.g.: large elastic extension. However, physical properties of
polymers depend not only on molecular weight and shape, but also on differences in
structure of the chains. It should be remembered that polymers are not usually of only one
distinctive structural type, though they are classified into different groups. Typical
polymer chain structures are: (a) linear, where mer units are joined together end to end in
single chains. E.g.: PVC, nylon. (b) branched, where side-branch chains are connected to
main ones. Branching of polymers lowers polymer density because of lower packing
efficiency. (c) cross-linked, where chains are joined one to another at various positions by
covalent bonds. This cross-linking is usually achieved at elevated temperatures by
additive atoms. E.g.: vulcanization of rubber. (d) network, trifunctional mer units with 3-
D networks comes under this category. E.g.: epoxies, phenol-formaldehyde.

2.4.3 Polymer crystallinity

Crystallinity in polymers is more complex than in metals. Polymer molecules are often
partially crystalline (semicrystalline), with crystalline regions dispersed within
amorphous material. The degree of crystallinity may range from completely amorphous
to almost entirely crystalline; on the other hand metals are almost always crystalline
whereas ceramics are either completely crystalline or noncrystalline. The degree of
crystallinity of a polymer depends on cooling path, and also on chain configuration. For
copolymers, the more irregular and random the mer arrangement, the greater is
probability for noncrystalline nature. Crystalline polymers are denser than amorphous
polymers, so the degree of crystallinity can be obtained from the measurement of density.

Different models have been proposed to describe the arrangement of molecules in

semicrytalline polymers. In the fringed-micelle model, the crystallites (micelles) are
embedded in an amorphous matrix. Polymer single crystals grown are shaped in regular
platelets (lamellae). Spherulites are chain-folded crystallites in an amorphous matrix that
grow radially in spherical shape grains. These are considered to be the polymer
analogue of grains in polycrystalline metals and ceramics. Many semicrystalline
polymers form spherulites; each spherulite consists of a collection of ribbonlike chain-
folded lamellar crystallites that radiate outward from its center. E.g.: polyethelene, PVC.

A polymers response to mechanical forces under elevated temperatures is related to its

molecular structure. Based on this response, polymers are classified as: thermoplasts
(soften when heated and harden when cooled), and thermosets (become permanently hard
when heat is applied and do not soften upon subsequent heating). Thermosets are
generally harder and stronger than thermoplasts, and have better dimensional stability.
Most of the cross-linked and network polymers are thermosets; whereas linear and some
branched polymers are thermoplasts.

2.4.4 Properties of polymers

Fluids and amorphous solids undergo viscous flow when external forces are applied. It is
well known that polymers exhibit very high viscosity in order of 1012 Pa.s at room
temperature. Polymers are non-Newtonian in nature, and formed into plastic products at a
temperature above their glass-transition temperature. It is evident that temperature has
very strong influence on mechanical behavior of polymers. Elastic strain occurs
simultaneously with viscous flow, resulting in visco-elastic deformation of polymers
under externally applied loads. Below the glass transition temperature elastic deformation
dominates and the material behaves rigid. In the range of glass temperature, the materials
is leathery; in the rubber plateau, polymers deform readily but quickly regain their
previous shape if the stress in removed. At still higher temperatures, under sustained
loads, the polymer deforms extensively by viscous flow. Figure below depicts
temperature effect on deformation behavior of polymers.

Figure 2.3: Dependence of polymer viscosity on temperature and/or loading rate.

2.5 Structure and properties of ceramics

Ceramics are inorganic and non-metallic materials that are commonly electrical and
thermal insulators, brittle and composed of more than one element (e.g., two in Al2O3).
As ceramics are composed of two or more elements, their crystal structures are generally
more complex than those of metals. Ceramic bonds are mixed, ionic and covalent, with a
proportion that depends on the particular ceramics. The ionic character is given by the
difference of electronegativity between the cations (+) and anions (-). Covalent bonds
involve sharing of valence electrons. Very ionic crystals usually involve cations which
are alkalis or alkaline-earths (first two columns of the periodic table) and oxygen or
halogens as anions.

The building criteria for the ceramic crystal structure are as follows:
- maintain neutrality (charge balance dictates chemical formula)
- achieve closest packing

The crystal stability condition i.e. condition of minimum energy implies maximum
attraction and minimum repulsion. This leads to contact and configurations such that
anions have the highest number of cation neighbors (coordination number) and vice
versa. The coordination number is dependent on cation-anion radius ratio, which can be
determined from geometric relations. Table 2.2 presents relevant coordination numbers
and radius ratios.

Table 2.2. Co-ordination number dependency on cation-anion radius ratio.

Cation-anion 0.155 0.225 0.414 0.732
< 0.155 > 1.000
radius ratio (rc/ra) 0.225 0.414 0.732 1.000
Coordination
2 3 4 6 8 12
number

Figure-2.4 presents schematic arrangement of cations and anions with respective

coordinate positions for different radius ratios.

Figure 2.4: Ion arrangements for different coordination numbers.

2.5.1 Ceramic crystal structures

AX-type ceramic crystal structures: Most common ceramics are made of equal number of
cations and anions, and are referred to as AX compounds (A-cation, and X-anion). These
ceramics assume many different structures, named after a common material that
possesses the particular structure.
Rock salt structure: here the coordination number is 6, i.e. rc/ra = 0.414-0.732. This
structure can be viewed as an FCC of anions with cations occupying center of each edge
and the center of the cell. Thus it can be said that lattice is made of two interpenetrating
FCC lattices, one composed of cations, and the other of anions. E.g.: NaCl, MgO, FeO.

Cesium Chloride structure: here the coordination number is 8. Crystal structure consists
of anions at corners of a cube while a cation occupies the center, and vice versa. E.g.:
CsCl.

Zinc Blende structure: here the coordination number is 4. Unit cell is composed of one
kind of ions occupying corners and face centers of a cube, while the other kind of ions
occupies the interior tetrahedral positions. E.g.: ZnS, SiC.

AmXp- type structures: when the charges of cation and anions are not the same, to
maintain the neutrality, ceramic structures with chemical formula would exist. For
example CaF2 with rc/ra = 0.8, and thus coordination number of 8. It can be expected
that crystal structure could be the same as that of CsCl. However cations are half many as
anions, thus only half the center positions are occupied. One unit cell shall be made of
eight cubes. E.g.: UO2, ThO2, PuO2.

AmBnXp-type structures: it is possible that ceramics do have more than one kind of
cations. E.g.: BaTiO3. unit cell is made of cube where Ba2+ ions occupies all eight
corners, T4+ occupies cube center, while O2- are at center of each face. This structure is
called perovskite crystal structure.

It is worth to understand and know more about some common most common ceramic in
nature. For example: silicates and carbon.

2.5.2 Silicates

Oxygen and Silicon are the most abundant elements in Earths crust. Their combination
(silicates) occurs in rocks, soils, clays and sand. The bond is weekly ionic, with Si4+ as
the cation and O2- as the anion. However the bonds in silicates are strongly of covalent
character with strong directional Si-O bonds. Basic unit of silicates structures are thus is
SiO44- tetrahedron that consists of four oxygen atoms at corners of tetrahedron, and
silicon atom at the center of it. Various silicate structures consists SiO44- unit bonded in
1-, 2-, and 3- dimensions.

In silica (SiO2) every oxygen atom is shared by adjacent tetrahedra. Silica can either be
crystalline (e.g., quartz) or amorphous, as in glass. Crystalline forms of silica are known
to be complicated and comparatively open, thus of low densities compared with
amorphous glasses. Soda glasses melt at lower temperature than amorphous SiO2 because
the addition of Na2O (soda) that act as network modifier breaks the tetrahedral network.
Addition of intermediates such as Al2O3, TiO2 substitute of silicon atoms and become
part of stabilized network. Addition of network modifiers and intermediates lowers
melting point, and thus it is easy to form glass, for instance, bottles.
In complicated silicate structure, corner oxygen atom of basic unit is shared by other
tetrahedra, resulting in formulas such as SiO44-, Si2O76-, Si3O96-, etc. The repeating unit of
2-D sheet or layered structure is represented as Si2O52-. Such layered structures are
characteristics of clays and other minerals.

2.5.3 Carbon

Carbon is not really a ceramic, but one of its allotropic form, diamond may be considered
as a ceramic. Diamond has very interesting and even unusual properties such as:
- possesses diamond-cubic structure (like Si, Ge)
- consists covalent C-C bonds
- having highest hardness of any material known
- very high thermal conductivity (unlike ceramics)
- transparent in the visible and infrared, with high index of refraction
- semiconductor (can be doped to make electronic devices)
- meta-stable (transforms to carbon when heated)

Synthetic diamonds are made by application of high temperatures and pressures or by

chemical vapor deposition. Future applications of this latter, cheaper production method
include hard coatings for metal tools, ultra-low friction coatings for space applications,
and microelectronics.

Graphite, another allotropic form of carbon, has a layered structure with very strong
hexagonal bonding within the planar layers (using 3 of the 3 bonding electrons) and
weak, van der Waals bonding between layers using the fourth electron. This leads to easy
inter-planar cleavage and applications as a lubricant and for writing (pencils). Graphite is
a good electrical conductor and chemically stable even at high temperatures. Applications
include furnaces, rocket nozzles, electrodes in batteries, etc.

Recently (1985) discovered allotropic form of carbon is the C60 molecule, also known as
fullerene or bucky-ball (after the architect Buckminster Fuller who designed the geodesic
structure that C60 resembles.). Structure of this form resembles a hallow spherical cluster
of 60 atoms, and is found to consist of 20 hexagons and 12 pentagons where no two
pentagons share a common edge. Fullerenes and related structures like nanotubes are
exceptionally stiff, strong, and ductile. Future applications of fullerenes are as a structural
material and possibly in microelectronics, due to the unusual properties that result when
fullerenes are doped with other atoms.

2.5.4 Imperfections in ceramics

Imperfections in ceramics include point defects and impurities. Their formation is

strongly affected by the condition of charge neutrality (creation of unbalanced charges
requires the expenditure of a large amount of energy). Both vacancies and interstitials are
possible in ceramics as in metals; however as ceramics have more then one element these
defects can be associated with each of these elements. Neutral charge defects include the
Frenkel and Schottky defects. A Frenkel-defect is a vacancy- interstitial pair of cations
(placing large anions in an interstitial position requires a lot of energy in lattice
distortion). A Schottky-defect is a pair of nearby cation and anion vacancies.

Non-stoichiometry refers to a change in composition so that the elements in the ceramic

are not in the proportion appropriate for the compound (condition known as
stoichiometry). To minimize energy, the effect of non-stoichiometry is a redistribution of
the atomic charges. Introduction of impurity atoms in the lattice is likely in conditions
where the charge is maintained. This is the case of electronegative impurities that
substitute lattice anions or electropositive substitutional impurities. This is more likely for
similar ionic radii since this minimizes the energy required for lattice distortion. Defects
will appear if the charge of the impurities is not balanced.

2.5.5 Mechanical response of ceramics

Brittle Fracture of Ceramics: The brittle fracture of ceramics limits their engineering
applications. It occurs due to the unavoidable presence of microscopic flaws (micro-
cracks, internal pores, and atmospheric contaminants) that result during cooling from the
melt. The flaws need to crack initiation, and crack propagation (perpendicular to the
applied stress) is usually trans-granular, along cleavage planes. The flaws cannot be
closely controlled in manufacturing; this leads to a large scatter in the fracture strength of
ceramic materials.

Under compressive stresses, however, flaws do not associate with amplification of stress.
Hence, the compressive strength of ceramics is typically ten times to their tensile
strength. This makes ceramics good structural materials under compression (e.g., bricks
in houses, stone blocks in the pyramids), but not in conditions of tensile stress, such as
under flexure.

Plastic deformation in crystalline ceramics is by slip, which is difficult due to the

structure and the strong local (electrostatic) potentials. There is very little plastic
deformation before fracture. Non-crystalline ceramics, like common glass, deform by
viscous flow (like very high-density liquids) as deformation can not occur by slip because
of absence of regular atomic structure, which occurs in metals. Characteristic property of
viscous flow, viscosity, is a measure of non-crystalline materials resistance to
deformation. It was found to decrease with increasing temperature. At room temperature,
the viscosity of many non-crystalline ceramics is extremely high.

References

1. William D. Callister, Jr, Materials Science and Engineering An introduction,

sixth edition, John Wiley & Sons, Inc. 2004.
2. B. D. Cullity and S. R. Stock, Elements of X-Ray Diffraction, Third Edition,
Prentice Hall, Upper Saddle River, NJ, 2001
3. Lawrence H. Van Vlack, Elements of Materials Science and Engineering, sixth
edition, Addison Wesley Longman, Inc. New York, 1998
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 3. Imperfections in Solids

3.1 Theoretical yield strength, Point defects, Line defects and Dislocations

3.1.1 Theoretical strength

Atoms are orderly arranged without any disruptions in ideal solids. Under mechanical
loads, all solids tend to reshape themselves. This occurs usually by process called slip,
the translation of one plane of atoms over another under implied shear stresses.

Consider two planes of atoms subjected a shear stress under applied load. Lets assume,
as shown in figure-3.1, distance between atoms in slip direction is b, and spacing between
the two planes is a.

Figure-3.1: Shear displacement of atom planes and variation of shear stress.

The shear stress is zero initially and also once planes moved one identity distance, b. In
between, it can be said that shear stress is a periodic function of displacement as each
atom is attracted toward the nearest atom of the other row. Thus the relation of shear
stress with displacement can be expressed as follows:

2 x
= m sin
b

where is the amplitude of sine wave, and b is the period. For small values of x/b, after
simplification,

2 x
m
b

It is safe to approximate at small values of displacement Hookes law is applicable. Thus,

Gx
= G =
a

Combing the above two equations, maximum shear stress at which slip should occur is

G b
m =
2 a

If it is assumed as b a, approximate value of shear stress needed for initiation of slip is

G
m =
2

Shear modulus, G, of solids is in the range of 20-150 GPa, and thus theoretical shear
strength could be in range 3-30 GPa. However, actual values of the shear stress required
for plastic deformation is only about 0.5-10 MPa i.e. about 100 times less than the
theoretical value. Because of this discrepancy, it must be said that a mechanism other
than shearing of planes must be operative under application of loads thus causing plastic
deformation.

3.1.2 Point defects

Defects exists any all solid materials. For ease of their characterization, defects are
classified on the basis of their geometry, which is realistic as defects are disrupted region
in a volume of a solid. Defects are:

a. Point defects (zero-dimensional)

b. Line defects (single dimensional)
c. Surface defects (two dimensional)
d. Volume defects (three dimensional)
Point defects, as the name implies, are imperfect point-like regions in the crystal. Typical
size of a point defect is about 1-2 atomic diameters. Different point defects are explained
in the following paragraphs. Figure-3.2 depicts various point defects.

A vacancy is a vacant lattice position from where the atom is missing. It is usually
created when the solid is formed by cooling the liquid. There are other ways of making a
vacancy, but they also occur naturally as a result of thermal excitation, and these are
thermodynamically stable at temperatures greater than zero. At equilibrium, the fraction
of lattice sites that are vacant at a given temperature (T) are:

n Q
= e kT
N

where n is the number of vacant sites in N lattice positions, k is gas or Boltzmanns

constant, T is absolute temperature in kelvins, and Q is the energy required to move an
atom from the interior of a crystal to its surface. It is clear from the equation that there is
an exponential increase in number of vacancies with temperature. When the density of
vacancies becomes relatively large, there is a possibility for them to cluster together and
form voids.

Figur-3.2: Schematic depiction of various point defects

An interstitial atom or interstitialcy is an atom that occupies a place outside the normal
lattice position. It may be the same type of atom as the rest surrounding it (self
interstitial) or a foreign impurity atom. Interstitialcy is most probable if the atomic
packing factor is low.

Another way an impurity atom can be fitted into a crystal lattice is by substitution. A
substitutional atom is a foreign atom occupying original lattice position by displacing the
parent atom.

In the case of vacancies and foreign atoms (both interstitial and substitutional), there is a
change in the coordination of atoms around the defect. This means that the forces are not
balanced in the same way as for other atoms in the solid, which results in lattice
distortion around the defect.
In ionic crystals, existence of point defects is subjected to the condition of charge
neutrality. There are two possibilities for point defects in ionic solids.

- when an ion displaced from a regular position to an interstitial position creating

a vacancy, the pair of vacancy-interstitial is called Frenkel defect. Cations are
usually smaller and thus displaced easily than anions. Closed packed structures
have fewer interstitials and displaced ions than vacancies because additional
energy is required to force the atoms into the interstitial positions.

- a pair of one cation and one anion can be missing from an ionic crystal, without
violating the condition of charge neutrality when the valency of ions is equal. The
pair of vacant sites, thus formed, is called Schottky defect. This type of point
defect is dominant in alkali halides. These ion-pair vacancies, like single
vacancies, facilitate atomic diffusion.

3.1.3 Line defects or Dislocations:

Line imperfections (one-dimensional defects) are also called Dislocations. They are
abrupt changes in the regular ordering of atoms along a line (dislocation line) in the solid.
They occur in high densities and strongly influence the mechanical properties of material.
They are characterized by the Burgers vector (b), whose direction and magnitude can be
determined by constructing a loop around the disrupted region and noticing the extra
inter-atomic spacing needed to close the loop. The Burgers vector in metals points in a
close packed lattice direction. It is unique to a dislocation.

Dislocations occur when an extra incomplete plane is inserted. The dislocation line is at
the end of the plane. Dislocations can be best understood by referring to two limiting
cases - Edge dislocation and Screw dislocation.

Edge dislocation or Taylor-Orowan dislocation is characterized by a Burgers vector that

is perpendicular to the dislocation line. It may be described as an edge of an extra plane
of atoms within a crystal structure. Thus regions of compression and tension are
associated with an edge dislocation. Because of extra incomplete plane of atoms, the
atoms above the dislocation line are squeezed together and are in state of compression
whereas atoms below are pulled apart and experience tensile stresses. Edge dislocation is
considered positive when compressive stresses present above the dislocation line, and is
represented by . If the stress state is opposite i.e. compressive stresses exist below the
dislocation line, it is considered as negative edge dislocation, and represented by . A
schematic view of edge dislocations are shown in figure 3.3.
 Figure-3.3: Edge dislocations.

A pure edge dislocation can glide or slip in a direction perpendicular to its length i.e.
along its Burgers vector in the slip plane (made of b and t vectors), on which dislocation
moves by slip while conserving number of atoms in the incomplete plane. It may move
vertically by a process known as climb, if diffusion of atoms or vacancies can take place
at appropriate rate. Atoms are added to the incomplete plane for negative climb i.e. the
incomplete plane increases in extent downwards, and vice versa. Thus climb motion is
considered as non-conservative, the movement by climb is controlled by diffusion
process.

Screw dislocation or Burgers dislocation has its dislocation line parallel to the Burgers
vector. A screw dislocation is like a spiral ramp with an imperfection line down its axis.
Screw dislocations result when displacing planes relative to each other through shear.
Shear stresses are associated with the atoms adjacent to the screw dislocation; therefore
extra energy is involved as it is in the case of edge dislocations. Screw dislocation is
considered positive if Burgers vector and t-vector or parallel, and vice versa. (t-vector
an unit vector representing the direction of the dislocation line). A positive screw
dislocation is represented by a dot surrounded by circular direction in clock-wise
direction, whereas the negative screw dislocation is represented by a dot
surrounded by a circular direction in anti-clock-wise direction. A schematic view of a
negative screw dislocation is shown in figure 3.4.
 Fifure-3.4: Negative screw dislocation.

A screw dislocation does not have a preferred slip plane, as an edge dislocation has, and
thus the motion of a screw dislocation is less restricted than the motion of an Edge
dislocation. As there is no preferred slip plane, screw dislocation can cross-slip on to
another plane, and can continue its glide under favorable stress conditions. However,
screw dislocation can not move by climb process, whereas edge dislocations can not
cross-slip.

Dislocations more commonly originate during plastic deformation, during solidification,

and as a consequence of thermal stresses that result from rapid cooling. Edge dislocation
arises when there is a slight mismatch in the orientation of adjacent parts of the growing
crystal. A screw dislocation allows easy crystal growth because additional atoms can be
added to the step of the screw. Thus the term screw is apt, because the step swings
around the axis as growth proceeds. Unlike point defects, these are not
thermodynamically stable. They can be removed by heating to high temperatures where
they cancel each other or move out through the crystal to its surface. Virtually all
crystalline materials contain some dislocations. The density of dislocations in a crystal is
measures by counting the number of points at which they intersect a random cross-
section of the crystal. These points, called etch-pits, can be seen under microscope. In an
annealed crystal, the dislocation density is the range of 108-1010 m-2.

Any dislocation in a crystal is a combination of edge and screw types, having varying
degrees of edge and screw character. Figure 3.5 depicts a schematic picture of a common
dislocation.
 Figure-3.5: Typical dislocation in a crystal.

As shown in figure 3.5, the Burgers vector will have the same magnitude and direction
over the whole length of dislocation irrespective of the character of the dislocation (edge,
screw, or mixed) i.e. Burgers vector is invariant. Other geometrical characteristics of
dislocations are:

- vectorial sum of Burgers vectors of dislocations meeting at a point, called node, must
be zero.

- t-vectors of all the dislocations meeting at a node must either point towards it or away
from it.

- a dislocation line can not end abruptly within the crystal. It can close on itself as a loop,
or ends either at a node or at the surface.

Dislocations have distortional energy associated with them as is evident from the
presence of tensile/compressive/shear stresses around a dislocation line. Strains can be
expected to be in the elastic range, and thus stored elastic energy per unit length of the
dislocation can be obtained from the following equation:

2
E Gb
2

where G shear modulus and b Burgers vector.

Dislocations in the real crystal can be classified into two groups based on their geometry
full dislocations and partial dislocations. Partial dislocations Burgers vector will be a
fraction of a lattice translation, whereas Burgers vector is an integral multiple of a lattice
translation for full dislocation. As mentioned above, elastic energy associated with a
dislocation is proportional to square of its Burgers vector; dislocation will tend to have
as small a Burgers vector as possible. This explains the reason for separation of
dislocations that tend to stay away from each other!
3.2 Interfacial defects, Bulk or Volume defects and Atomic vibrations

3.2.1 Interfacial defects

Interfacial defects can be defined as boundaries that have two dimensional imperfections
in crystalline solids, and have different crystal structures and/or crystallographic
orientations on either side of them. They refer to the regions of distortions that lie about a
surface having thickness of a few atomic diameters. For example: external surfaces, grain
boundaries, twin boundaries, stacking faults, and phase boundaries. These imperfections
are not thermodynamically stable, rather they are meta-stable imperfections. They arise
from the clustering of line defects into a plane.

External surface: The environment of an atom at a surface differs from that of an atom in
the bulk; especially the number of neighbors (coordination) at surface is less. Thus the
unsaturated bonds of surface atoms give rise to a surface energy. This result in relaxation
(the lattice spacing is decreased) or reconstruction (the crystal structure changes). To
reduce the energy, materials tend to minimize, if possible, the total surface area.

Grain boundaries: Crystalline solids are, usually, made of number of grains separated by
grain boundaries. Grain boundaries are several atoms distances wide, and there is
mismatch of orientation of grains on either side of the boundary as shown in figure-3.6.
When this misalignment is slight, on the order of few degrees (< 10), it is called low
angle grain boundary. These boundaries can be described in terms of aligned dislocation
arrays. If the low grain boundary is formed by edge dislocations, it is called tilt boundary,
and twist boundary if formed of screw dislocations. Both tilt and twist boundaries are
planar surface imperfections in contrast to high angle grain boundaries. For high angle
grain boundaries, degree of disorientation is of large range (> 15). Grain boundaries are
chemically more reactive because of grain boundary energy. In spite of disordered
orientation of atoms at grain boundaries, polycrystalline solids are still very strong as
cohesive forces present within and across the boundary.

Figure-3.6: Schematic presentation of grain boundaries.

Twin boundaries: It is a special type of grain boundary across which there is specific
mirror lattice symmetry. Twin boundaries occur in pairs such that the orientation change
introduced by one boundary is restored by the other (figure-3.7). The region between the
pair of boundaries is called the twinned region. Twins which forms during the process of
recrystallization are called annealing twins, whereas deformation twins form during
plastic deformation. Twinning occurs on a definite crystallographic plane and in a
specific direction, both of which depend on the crystal structure. Annealing twins are
typically found in metals that have FCC crystal structure (and low stacking fault energy),
while mechanical/deformation twins are observed in BCC and HCP metals. Annealing
twins are usually broader and with straighter sides than mechanical twins. Twins do not
extend beyond a grain boundary.

Figure-3.7: A pair of twin boundaries.

Stacking faults: They are faults in stacking sequence of atom planes. Stacking sequence
in an FCC crystal is ABC ABC ABC , and the sequence for HCP crystals is AB AB
AB. When there is disturbance in the stacking sequence, formation of stacking faults
takes place. Two kinds of stacking faults in FCC crystals are: (a) ABC AC ABCwhere
CA CA represent thin HCP region which is nothing but stacking fault in FCC, (b) ABC
ACB CABC is called extrinsic or twin stacking fault. Three layers ACB constitute the
twin. Thus stacking faults in FCC crystal can also be considered as submicroscopic twins.
This is why no microscopic twins appear in FCC crystals as formation of stacking faults
is energetically favorable. Stacking fault energy varies in range 0.01-0.1 J/m2. Lower the
stacking fault energy, wider the stacking fault, metal strain hardens rapidly and twin
easily. Otherwise, metals of high stacking fault energy i.e. narrower stacking faults show
a deformation structure of banded, linear arrays of dislocations.

Phase boundaries exist in multiphase materials across which there is sudden change in
physical/chemical characteristics.

3.2.2 Bulk or Volume defects

Volume defects as name suggests are defects in 3-dimensions. These include pores,
cracks, foreign inclusions and other phases. These defects are normally introduced during
processing and fabrication steps. All these defects are capable of acting as stress raisers,
and thus deleterious to parent metals mechanical behavior. However, in some cases
foreign particles are added purposefully to strengthen the parent material. The procedure
is called dispersion hardening where foreign particles act as obstacles to movement of
dislocations, which facilitates plastic deformation. The second-phase particles act in two
distinct ways particles are either may be cut by the dislocations or the particles resist
cutting and dislocations are forced to bypass them. Strengthening due to ordered particles
is responsible for the good high-temperature strength on many super-alloys. However,
pores are detrimental because they reduce effective load bearing area and act as stress
concentration sites.

3.2.3 Atomic vibrations

Atomic vibrations occur, even at zero temperature (a quantum mechanical effect) and
increase in amplitude with temperature. In fact, the temperature of a solid is really just a
measure of average vibrational activity of atoms and molecules. Vibrations displace
transiently atoms from their regular lattice site, which destroys the perfect periodicity. In
a sense, these atomic vibrations may be thought of as imperfections or defects. At room
temperature, a typical vibrational frequency of atoms is of the order of 1013 vibrations per
second, whereas the amplitude is a few thousandths of a nanometer. Many properties and
processes in solids are manifestations of this vibrational atomic motion. For example:
melting occurs once the atomic bonds are overcome by vigorous vibrations.

References

1. ASM handbook, Metallography and Microstructures, Vol. 9, ASM International,

Materials Park, OH, 1985
2. William D. Callister, Jr, Materials Science and Engineering An introduction,
sixth edition, John Wiley & Sons, Inc. 2004.
3. V. Raghavan, Materials Science and Engineering, third edition, Prentice Hall of
India Private Limited, New Delhi, 1990.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 4. Mechanical Properties of Metals

Most of the materials used in engineering are metallic in nature. The prime reason simply
is the versatile nature of their properties those spread over a very broad range compared
with other kinds of materials. Many engineering materials are subjected to forces both
during processing/fabrication and in service. When a force is applied on a solid material,
it may result in translation, rotation, or deformation of that material. Aspects of material
translation and rotation are dealt by engineering dynamics. We restrict ourselves here to
the subject of material deformation under forces. Deformation constitutes both change in
shape, distortion, and change in size/volume, dilatation. Solid material are defined such
that change in their volume under applied forces in very small, thus deformation is used
as synonymous to distortion. The ability of material to with stand the applied force
without any deformation is expressed in two ways, i.e. strength and hardness. Strength is
defined in many ways as per the design requirements, while the hardness may be defined
as resistance to indentation of scratch.

Material deformation can be permanent or temporary. Permanent deformation is

irreversible i.e. stays even after removal of the applied forces, while the temporary
deformation disappears after removal of the applied forces i.e. the deformation is
recoverable. Both kinds of deformation can be function of time, or independent of time.
Temporary deformation is called elastic deformation, while the permanent deformation is
called plastic deformation. Time dependent recoverable deformation under load is called
anelastic deformation, while the characteristic recovery of temporary deformation after
removal of load as a function of time is called elastic aftereffect. Time dependent i.e.
progressive permanent deformation under constant load/stress is called creep. For visco-
elastic materials, both recoverable and permanent deformations occur together which are
time dependent. When a material is subjected to applied forces, first the material
experiences elastic deformation followed by plastic deformation. Extent of elastic- and
plastic- deformations will primarily depend on the kind of material, rate of load
application, ambient temperature, among other factors. Change over from elastic state to
plastic state is characterized by the yield strength (0) of the material.

Forces applied act on a surface of the material, and thus the force intensity, force per unit
area, is used in analysis. Analogous to this, deformation is characterized by percentage
change in length per unit length in three distinct directions. Force intensity is also called
engineering stress (or simply stress, s), is given by force divided by area on which the
force is acting. Engineering strain (or simply strain, e) is given by change in length
divided by original length. Engineering strain actually indicates an average change in
length in a particular direction. According to definition, s and e are given as

P L L0
s= ,e =
A0 L0

where P is the load applied over area A, and as a consequence of it material attains the
final length L from its original length of L0.

Because material dimensions changes under application of the load continuously,

engineering stress and strain values are not the true indication of material deformation
characteristics. Thus the need for measures of stress and strain based on instantaneous
dimensions arises. Ludwik first proposed the concept of, and defined the true strain or
natural strain () as follows:

L1 L0 L2 L1 L3 L2
= + + + ...
L0 L1 L2

L
dL L
=
L0
L
= ln
L0

As material volume is expected to be constant i.e. A0L0=AL, and thus

L A
= ln = ln 0 = ln(e + 1)
L0 A

There are certain advantages of using true strain over conventional strain or engineering
strain. These include (i) equivalent absolute numerical value for true strains in cases of
tensile and compressive for same intuitive deformation and (ii) total true strain is equal to
the sum of the incremental strains. As shown in figure-4.1, if L1=2 L0 and L2=1/2 L1=L0,
absolute numerical value of engineering strain during tensile deformation (1.0) is
different from that during compressive deformation (0.5). However, in both cases true
strain values are equal (ln [2]).

True stress () is given as load divided by cross-sectional area over which it acts at an
instant.

P P A0
= = = s(e + 1)
A A0 A
It is to be noted that engineering stress is equal to true stress up to the elastic limit of the
material. The same applies to the strains. After the elastic limit i.e. once material starts
deforming plastically, engineering values and true values of stresses and strains differ.
The above equation relating engineering and true stress-strains are valid only up to the
limit of uniform deformation i.e. up to the onset of necking in tension test. This is
because the relations are developed by assuming both constancy of volume and
homogeneous distribution of strain along the length of the tension specimen. Basics of
both elastic and plastic deformations along with their characterization will be detailed in
this chapter.

4.1 Elastic deformation and Plastic deformation

4.1.1 Elastic deformation

Elastic deformation is reversible i.e. recoverable. Up to a certain limit of the applied

stress, strain experienced by the material will be the kind of recoverable i.e. elastic in
nature. This elastic strain is proportional to the stress applied. The proportional relation
between the stress and the elastic strain is given by Hookes law, which can be written as
follows:

= E

where the constant E is the modulus of elasticity or Youngs modulus,

Though Hookes law is applicable to most of the engineering materials up to their elastic
limit, defined by the critical value of stress beyond which plastic deformation occurs,
some materials wont obey the law. E.g.: Rubber, it has nonlinear stress-strain
relationship and still satisfies the definition of an elastic material. For materials without
linear elastic portion, either tangent modulus or secant modulus is used in design
calculations. The tangent modulus is taken as the slope of stress-strain curve at some
specified level, while secant module represents the slope of secant drawn from the origin
to some given point of the - curve, as shown in figure-4.1.
 Figure-4.1: Tangent and Secant moduli for non-linear stress-strain relation.

If one dimension of the material changed, other dimensions of the material need to be
changed to keep the volume constant. This lateral/transverse strain is related to the
applied longitudinal strain by empirical means, and the ratio of transverse strain to
longitudinal strain is known as Poissons ratio (). Transverse strain can be expected to
be opposite in nature to longitudinal strain, and both longitudinal and transverse strains
are linear strains. For most metals the values of are close to 0.33, for polymers it is
between 0.4 0.5, and for ionic solids it is around 0.2.

Stresses applied on a material can be of two kinds normal stresses, and shear stresses.
Normal stresses cause linear strains, while the shear stresses cause shear strains. If the
material is subjected to torsion, it results in torsional strain. Different stresses and
corresponding strains are shown in figure-4.2.
 Figure 4.2: Schematic description of different kinds of deformations/strains.

Analogous to the relation between normal stress and linear strain defined earlier, shear
stress () and shear strain () in elastic range are related as follows:

= G

where G is known as Shear modulus of the material. It is also known as modulus of

elasticity in shear. It is related with Youngs modulus, E, through Poissons ratio, , as

E
G=
2(1 + )

Similarly, the Bulk modulus or volumetric modulus of elasticity K, of a material is defined

as the ratio of hydrostatic or mean stress (m) to the volumetric strain (). The relation
between E and K is given by

m E
K= =
3(1 2 )

Let x, y and z are linear stresses and x, y and z are corresponding strains in X-, Y-
and Z- directions, then

x + y +z
m =
3

Volumetric strain or cubical dilatation is defined as the change in volume per unit
volume.

= (1 + x )(1 + y )(1 + z ) 1 x + y + z , = 3 m

where m is mean strain or hydrostatic (spherical) strain defined as

x +y +z
m =
3

An engineering material is usually subjected to stresses in multiple directions than in just

one direction. If a cubic element of a material is subjected to normal stresses x, y, and
z, strains in corresponding directions are given by

x=
1
E
[ ] 1
[ ] 1
[
x ( y + z ) , y= y ( x + x ) , and z = z ( x + y )
E E
]
 E
and x + y + z = ( x + y + z )
1 2

The strain equation can be modified as

1 +
x=
1
E
[ ]
x ( y + z ) =
E
x ( x + y + z )
E

After substituting the sum-of-stress into the above equation, stress can be related to
strains as follows:

E E E
x = x + ( x + y + z ) = x + ( x + y + z ) = 2G x +
1 + (1 + )(1 2 ) 1 +

where is called Lames constant.

E
=
(1 + )(1 2 )

Using the above equations, it is possible to find strains from stresses and vice versa in
elastic range.

The basis for elastic deformation is formed by reversible displacements of atoms from
their equilibrium positions. On an atomic scale, elastic deformation can be viewed as
small changes in the inter-atomic distances by stretching of inter-atomic bonds i.e. it
involve small changes in inter-atomic distances. Elastic moduli measure the stiffness of
the material. They are related to the second derivative of the inter-atomic potential, or the
first derivative of the inter-atomic force vs. inter-atomic distance (dF/dr) (figure-4.3). By
examining these curves we can tell which material has a higher modulus.

Elastic modulus can also be said as a measure of the resistance to separation of adjacent
atoms, and is proportional to the slope of the inter-atomic fore Vs inter-atomic distance
curve (figure-4.3). Hence values of modulus of elasticity are higher for ceramic materials
which consist of strong covalent and ionic bonds. Elastic modulus values are lower for
metal when compared with ceramics, and are even lower for polymers where only weak
covalent bonds present.

Moreover, since the inter-atomic forces will strongly depend on the inter-atomic distance
as shown in figure-4.3, the elastic constants will vary with direction in the crystal lattice
i.e. they are anisotropic in nature for a single crystal. However, as a material consists of
number of randomly oriented crystals, elastic constants of a material can be considered as
isotropic.
Figure-4.3: Graph showing variation of inter-atomic forces against inter-atomic distance
for both weak and strong inter-atomic bonds.

The elastic moduli are usually measured by direct static measurements in tension or
torsion tests. For more precise measurements, dynamic techniques are employed. These
tests involve measurement of frequency of vibration or elapsed time for an ultrasonic
pulse to travel down and back in a specimen. Because strain cycles occur at very high
rates, no time for heat transfer and thus elastic constants are obtained under adiabatic
conditions. Elastic modulus obtained under isothermal and adiabatic conditions are
related as follows:

Eiso
E adiabatic =
EisoT 2
1
9c

where is volume coefficient of thermal expansion, and c is the specific heat. It can be
observed that with increasing temperature, the modulus of elasticity diminishes. This is
because, intensity of thermal vibrations of atoms increases with temperature which
weakens the inter-atomic bonds.

4.1.2 Plastic deformation

When the stress applied on a material exceeds its elastic limit, it imparts permanent non-
recoverable deformation called plastic deformation in the material. Microscopically it can
be said of plastic deformation involves breaking of original atomic bonds, movement of
atoms and the restoration of bonds i.e. plastic deformation is based on irreversible
displacements of atoms through substantial distances from their equilibrium positions.
The mechanism of this deformation is different for crystalline and amorphous materials.
For crystalline materials, deformation is accomplished by means of a process called slip
that involves motion of dislocations. In amorphous materials, plastic deformation takes
place by viscous flow mechanism in which atoms/ions slide past one another under
applied stress without any directionality.

Plastic deformation is, as elastic deformation, also characterized by defining the relation
between stresses and the corresponding strains. However, the relation isnt simpler as in
case of elastic deformation, and in fact it is much more complex. It is because plastic
deformation is accomplished by substantial movement of atomic planes, dislocations
which may encounter various obstacles. This movement becomes more complex as
number slip systems may get activated during the deformation.

The analysis of plastic deformation, and the large plastic strains involved is important in
many manufacturing processes, especially forming processes. It is very difficult to
describe the behavior of metals under complex conditions. Therefore, certain simplifying
assumptions are usually necessary to obtain an amenable mathematical solution.
Important assumptions, thus, involved in theory of plasticity are neglecting the following
aspects:

(i) Anelastic strain, which is time dependent recoverable strain.

(ii) Hysteresis behavior resulting from loading and un-loading of material.

(iii) Bauschinger effect dependence of yield stress on loading path and direction.

The relations describing the state of stress and strain are called constitutive equations
because they depend on the material behavior. These relations are applicable to any
material whether it is elastic, plastic or elastic-plastic. Hookes law which states that
strain is proportional to applied stress is applicable in elastic range where deformation is
considered to be uniform. However, plastic deformation is indeed uniform but only up to
some extent of strain value, where after plastic deformation is concentrated the
phenomenon called necking. The change over from uniform plastic deformation to non-
uniform plastic deformation is characterized by ultimate tensile strength (u). As a result
of complex mechanism involved in plastic deformation and its non-uniform distribution
before material fractures, many functional relations have been proposed to quantify the
stress-strain relations in plastic range.

A true stress-strain relation plotted as a curve is known as flow curve because it gives the
stress () required to cause the material to flow plastically to any given extent of strain ()
under a set of conditions. Other important parameters affecting the stress-strain curve are:
rate at which the load is applied / strain rate ( & ), and the temperature of the material (T,
in K) i.e.

= fn( , &, T , microstruc ture )

Following are the most common equations that describe the material flow behavior:

= K n
where K is strength coefficient, and n is strain hardening exponent. The strain-
hardening exponent may have values from n=0 (perfectly plastic solid) to n=1 (elastic
solid). For most metals n has values between 0.10 and 0.50. The Power law equation
described above is also known as Holloman-Ludwig equation. Another kind of power
equation which describes the material behavior when strain rate effect is prominent:

= K& m

where m is the index of strain-rate sensitivity. If m=0, the stress is independent of strain
rate. m=0.2 for common metals. If m=0.4-0.9, the material may exhibit super-plastic
behavior ability to deform by several hundred percent of strain without necking. If m=1,
the material behaves like a viscous liquid and exhibits Newtonian flow.

Deviations from the above power equations are frequently observed, especially at low
strains (<10-3) or high strains (>>1,0). One common type of deviation is for a log-log
plot of Power equation to result in two straight lines with different slopes. For data which
do not follow the Power equation, following equation could be used

= K ( 0 + ) n

where 0 - is the strain material had under gone before the present characterization.

Another common variation from Power law equation, also known as Ludwig equation, is:

= o + K n

where o - is the yield strength of the material.

Many other expressions for flow curve are available in literature. The true strain used in
the above equations should actually be the plastic strain value, given by the following
equation, where elastic strains can be safely neglected because plastic strains are very
much when compared with elastic strain values.

0
p = e =
E

Both temporary elastic deformation and permanent plastic deformation are compared in
tabel-4.1.

Table-4.1: Elastic deformation Vs Plastic deformation.

Elastic deformation Plastic deformation

Reversible Not reversible
Depends on initial and final states of Depends on loading path
 stress and strain
Stress is proportional to strain
No strain hardening effects
No simple relation between stress
and strain
Strain hardening effects

4.2 Interpretation of tensile stress-strain curves

It is well known that material deforms under applied loads, and this deformation can be
characterized by stress-strain relations. The stress-strain relation for a material is usually
obtained experimentally. Many kinds of experiments those differ in way of loading the
material are standardized. These include tension test, compression test (upsetting), plane
strain compression test, torsion test, etc.

The engineering tension test is commonly used to provide basic design information on
the strength characteristics of a material. Standardized test procedure is explained by
ASTM standard E0008-04. In this test a specimen is subjected to a continually increasing
uni-axial tensile force while measuring elongation simultaneously. A typical plot of load-
elongation is given in the figure-4.4. The curve also assumes the shape of an engineering
stress engineering strain curve after dividing the load with initial area and the
elongation with initial length. The two curves are frequently used interchangeably as they
differ only by constant factors. The shape and relative size of the engineering stress-strain
curve depends on material composition, heat treatment, prior history of plastic
deformation, strain rate, temperature and state of stress imposed on specimen during the
test.

Figure-4.4: Typical load elongation / engineering stress engineering strain / true

stress true strain curve.
 Figure-4.5: Magnified view of initial part of stress-strain curve.

As shown in figure-4.4, at initial stages load is proportional to elongation to a certain

level (elastic limit), and then increases with elongation to a maximum (uniform plastic
deformation), followed by decrease in load due to necking (non-uniform plastic
deformation) before fracture of the specimen occurs. Magnified view of the initial stage
of the curve is shown in figure-4.5. Along the segment AB of the curve, engineering
stress is proportional to engineering strain as defined by Hookes law, thus point-B is
known as proportional limit. Slope of the line AB gives the elastic modulus of the
material. With further increase in stress up to point-C, material can still be elastic in
nature. Hence point-C is known as elastic limit. It is to be noted that there is no point
where exactly material starts deform plastically i.e. there is no sharp point to indicate start
of the yield. Otherwise point-C can also be called yield point. Thus it is common to
assume that stress value at 0.2% offset strain as yield strength (0), denoted by point-D.
This offset yield strength is also called proof stress. Proof stress is used in design as it
avoids the difficulties in measuring proportional or elastic limit. For some materials
where there is essentially no initial linear portion, offset strain of 0.5% is frequently used.
In figure-4.5 distance between points-B, C and D is exaggerated for clarity. For many
materials it is difficult to make any difference between points-B and C, and point-D also
coincides with point-B/C. After the yield point, point-D, stress reaches a maximum at
point-E (tensile strength (f) as figure-4.4) till where plastic deformation is uniform along
the length of the specimen. Stress decreases hereafter because of onset of necking that
result in non-uniform plastic deformation before specimen fractures at point-F, fracture
limit. Yield strength and tensile strength are the parameters that describe the materials
strength, while percent elongation and reduction in cross-sectional area are used to
indicate the materials ductility extent of material deformation under applied load
before fracture.

Percent elongation (e) and reduction is area (r) are related as follows:

Volume constancy A0 L0 = AL
 L L0
Percent elongation e=
L0

A0 A A0 1
Reduction in area r= , =
A0 A 1 r

L L0 L A 1 r
Relation e= = 1 = 0 1 = 1 =
L0 L0 A 1 r 1 r

Other important parameters from the engineering stress-strain curve are resilience and
toughness. Resilience is defined as ability of a material to absorb energy when deformed
elastically and to return it when unloaded. Approximately this is equal to the area under
elastic part of the stress-strain curve, and equal to area-ADH in figure-4.5. This is
measured in terms of modulus of resilience (Ur) strain energy per unit volume required
to stress the material from zero stress to yield stress (0 = s0).

1 1 s s2
Ur = s 0 e0 = s 0 0 = 0
2 2 E 2E

where e0 it elastic strain limit. Toughness (Ut) of the materials is defined as its ability to
absorb energy in the plastic range. In other terms, it can be said to equal to work per unit
volume which can be done on the material without causing it to rupture. It is considered
that toughness of a material gives an idea about both strength and ductility of that
material. Experimentally toughness is measured by either Charpy or Izod impact tests.
Numerically the toughness value is equal to the area under the stress-strain curve, area-
AEFI, approximately of rectangular shape.

s 0 + su
U t su e f ef
2

For brittle materials (those have high elastic modulus and low ductility), stress-strain
curve is considered to assume the shape of parabola, thus

2
Ut su e f
3

where su ultimate tensile strength and ef strain at fracture.

As explained in earlier section, engineering stress-strain curve is not true representative

of the material behavior. But, the flow curve (true stress-true strain curve) represents the
basic plastic-flow characteristics of the material. More upon, special feature of the flow
curve is that any point on the curve can be considered as yield point i.e. if load is
removed and then reapplied, material will behave elastically throughout the entire range
of reloading. It can be said from the relations between engineering stress-strain and true
stress-strains the true stress-true strain curve is always to the left of the engineering curve
until the maximum load is reached. Point-E on represents the corresponding location on
true stress-strain curve to ultimate tensile stress point-E on engineering stress-strain
curve. After the point-E, flow curve is usually linear up to fracture, and in some cases its
slope decreases continually up to fracture.

In correspondence to different variables defined based on engineering stress-strain curve,

following parameters are defined based on flow curve:

Pmax P A A
True stress at maximum load u = , su = max , u = ln 0 u = su 0 = su e u
Au Ao Au Au

A0 1
True fracture strain f = ln = ln
Af 1 r

It is not possible to calculate f from estimated values of ef because the

relation between them is not valid beyond the onset of necking.

A0
True uniform strain u = ln
Au

The uniform strain is useful in estimating the formability of metals from

the results of a tension test. The condition u = n represents the onset of
necking.

Au
True local necking strain n = ln
Af

It is important to note that the rate of strain hardening (d/d) is not identical to the strain-
hardening exponent (n). They differ as follows:

d (log ) d (ln ) d d
n= = = or =n
d (log ) d (ln ) d d

4.3 Yielding under multi-axial stress, Yield criteria, Macroscopic aspects of plastic
deformation and Property variability & Design considerations

4.3.1 Yielding under multi-axial stress

Once the necking starts to form i.e. material starts to deform plastically but in non-
uniform mode, uni-axial state of stress turns into multi-axial (tri-axial) stress state. The
necked region is in effect a mild notch. The chief effect of the notch is not in introducing
a stress concentration but in producing a tri-axial state of stress i.e. introduction of
transverse stresses. As a result of tri-axial state of stress, yield stress becomes greater than
the uni-axial yield stress, 0, because it is more difficult to spread the yielded zone in the
presence of tri-axial stresses. Thus, the average true stress at the neck is higher than the
stress which would be required to cause flow in simple tension prevailed.

Bridgman put forwarded mathematical analysis to calculate the true stress from measured
stress in axial direction under tri-axial stress condition. His analysis is based on the
following assumptions: counter of the neck is approximated by the arc of a circle; cross-
section of the necked region remains circular; von Mises yield criterion is applicable;
strains are constant over the cross-section of the neck. Bridgmans correction is
applicable from the onset of necking for flow curve i.e. from point-E shown in figure-
4.4. Corrected yield stress under tri-axial state of stress is given as follows:

( x ) avg
=
(1 + 2 R / a)[ln(1 + a / 2 R)]

where (x)avg measured stress in the axial direction, a smallest radius in the neck region,
R radius of the curvature of neck (figure-4.6).

Figure-4.6: Geometry of necked region in cylindrical specimen under tensile load.

4.3.2 Yield criteria

It is known that material yields under condition of applied stress(es). In uni-axial loading,
like in a tension test, yield occurs i.e. macroscopic plastic flow starts at yield stress, 0.
However the situation is much more complicated in presence of multi-axial stresses. It
could be expected that yielding condition under multi-axial stresses to be a function of
particular combination of principal stresses.
Presently available yield criteria are essentially empirical relations. Thus, it needs to
satisfy some experimental observations. These include: hydrostatic component of stress
state should not influence the stress where the yield occurs; it must be an invariant
function i.e. independent of the choice of axes. These lead to the statement that yield
criterion must be some function of invariants of stress deviator.

There are two generally accepted criteria are in use for predicting the onset of yielding in
ductile materials: von Mises or Distortion-energy criterion and Maximum-shear-stress or
Tresca criterion.

von Mises or Distortion-energy criterion: it states that yielding occur when the second
invariant of the stress deviator J2 exceeded some critical value.

J2 = k 2

where J 2 =
1
6
[ ]
( 1 2 ) 2 + ( 2 3 ) 2 + ( 3 1 ) 2 ; 1, 2 and 3 are principal stresses.

It implies that yield condition in not dependent on any particular stress, but instead it
depends on all three principal stresses. And as the criterion is based on differences of
normal stresses, it is independent of hydrostatic stress component. In energy terms, it can
be said that yielding occurs when the distortion energy reaches a critical value. Distortion
energy is part of total strain energy per unit volume that is involved in change of shape as
opposed to a change in volume.

To evaluate the constant k, lets consider the yield in uni-axial tension test i.e. 1 = 0,
2= 3= 0. Thus,

1 2
( 0 + 02 ) = k 2 0 = 3k
6
0 =
1
[
( 1 2 ) 2 + ( 2 3 ) 2 + ( 3 1 ) 2 ]
1
2

To identify the constant k, by considering the state of stress in pure shear (torsion test): 1
=- 3 = , 2 = 0.

1 2
( 1 + 12 + 4 12 ) = k 2 1 = k
6

i.e. k represents the yield stress under pure shear, whereas 0 represents the yield stress
under uni-axial tension. These two yield stresses can be related as follows:

1
k= 0 = 0.577 0
3
von Mises yield criterion can also be interpreted as yielding occurs if octahedral shear
stress reaches a critical value. This is shear stress on octahedral plane which makes equal
angles with all three principal axes. It also represents the mean square of the shear stress
averaged over all orientations in the solid.

Cosine of angle between normal to a face of octahedron and a nearest principal axis is
1/3, i.e. the angle is 54 44. Normal octahedral stress (oct) is equivalent to hydrostatic
component of the stress system. Thus it can result in yielding, but shear octahedral stress
(oct) do.

1 + 2 + 3
oct = m =
3

oct =
1
3
[
( 1 2 ) 2 + ( 2 3 ) 2 + ( 3 1 ) 2 ]1 2

2
oct = 0 = 0.471 0
3

Corresponding octahedral strains are given as follows:

1 + 2 + 3
oct =
3
, =
2
3
[
( 1 2 ) 2 + ( 2 3 ) 2 + ( 3 1 ) 2 ]
1 2

Maximum-shear-stress or Tresca criterion: this criterion states that yielding occurs

once the maximum shear stress of the stress system reaches the value of shear stress in
uni-axial tension test.

If 1, 2 and 3 are principal stresses arranged in descending order, maximum shear stress
is given as

1 3
max =
2

Though this criterion is simpler than the von Mises yield criterion, here it is necessary to
know before-hand the maximum and minimum principal stresses from the stress system.

As with the earlier criterion, under uni-axial tension test conditions (1 = 0, 2= 3= 0),

1 3 0
max = =0 = 1 3 = 0
2 2

Under pure shear stress conditions (1 =- 3 = k, 2 = 0),

 1 3 1
k= = 0
2 2

Thus the other mathematical form for the criterion can be

1 3 = 1' 31 = k

Note that the yield locus for the Tresca criterion falls inside of the von Mises yield
ellipse. The two yielding criteria predict the same yield stress for conditions of uni-axial
stress and balanced bi-axial stress (1 = 3). The greatest divergence between the two
criteria occurs for pure shear (1 = - 3) where yield stress from von Mises criterion
1 1
( 0 ) is 15.5% greater than the yield stress from Tresca criterion ( 0 ).
3 2

4.3.3 Macroscopic aspects of plastic deformation

In single crystal, plastic deformation is accomplished by the process called slip, and
sometimes by twinning. The extent of slip depends on many factors including external
load and the corresponding value of shear stress produced by it, the geometry of crystal
structure, and the orientation of active slip planes with the direction of shearing stresses
generated. Schmid first recognized that single crystals at different orientations but of
same material require different stresses to produce slip. The dependence of various
factors has been summarized using a parameter critical resolved shear stress, R, given
as

P cos P
= = cos cos = cos cos
A cos A
m = cos cos

where P external load applied, A cross-sectional area over which the load applied,
angle between slip direction and tensile axis, angle between normal to the slip plane
and the tensile axis and m Schmid factor.

Shear stress is maximum for the condition where = = 45 . If either of the angles are
equal to 90 , resolved shear stress will be zero, and thus no slip occurs. If the conditions
are such that either of the angles is close to 90 , crystal will tend to fracture rather than
slip. Single crystal metals and alloys are used mainly for research purpose and only in a
few cases of engineering applications.

Almost all engineering alloys are polycrystalline. Gross plastic deformation of a

polycrystalline specimen corresponds to the comparable distortion of the individual
grains by means of slip. This involves generation, movement and (re-)arrangement of
dislocations. During deformation, mechanical integrity and coherency are maintained
along the grain boundaries; that is, the grain boundaries are constrained, to some degree,
in the shape it may assume by its neighboring grains. Lets assume that the grains are
equi-axed before deformation, or have approximately the same dimension in all
directions. If the material is imparted with deformation in a particular direction, the
grains become elongated along the direction in which the specimen was extended, as
shown in figure-4.7.

Another noticeable aspect of plastic deformation is arrangement of dislocations forming

cell like configurations with clear areas in the centers of the cells. With increasing
deformation, the cell structure becomes denser. In early stages of deformation, slip is
essentially confined to primary glide planes and thus dislocations form coplanar arrays.
As deformation proceeds, cross slip comes into picture, leading to tangled network of
dislocations forming cell walls, figure-4.8. Cell size decreases with increase in
deformation but reaches a fixed size depending on the material, the strain, the strain rate
and the temperature of deformation. In materials with low stacking fault energy where
cross-slip id difficult, development of cell structure is less pronounced.

Figure-4.7: Grain elongation under applied external forces.

Figure-4.8: Schematic view of cell structure.

Schematic presentation of macroscopic observation of a tensile specimen is shown in

figure-4.9. As explained earlier, initially specimen undergoes elastic deformation,
followed by uniform plastic deformation. After attaining maximum load capacity, non-
uniform plastic deformation sets-in. this leads to formation of macroscopic defects, like
cracks, which causes eventual fracture failure of the sample.

Figure-4.9: Schematic macroscopic presentation of plastic deformation in tensile

specimen.

4.3.4 Property variability & Design consideration

Scatter in measured properties of engineering materials is inevitable because of number

of factors such as test method, specimen fabrication procedure, operator bias, apparatus
calibration etc. In spite of property variation, some typical value is desirable. Most
commonly used typical value is by taking an average of the data. The average ( x ) of the
a parameter, xi, is given as
n

x i
x= i =1

where n is the number of samples, and xi is the discrete measurement.

In some instances, it is desirable to have an idea about the degree of variability, scatter, of
the measured data. Most common measure of this degree of variability is the standard
deviation, s, which is given by
 1 2
n 2
( xi x )
s = i =1
n 1

A large value for s means a high degree of scatter. Scatter is usually represented in
graphical form using error bars. If a parameter is averaged to x , and the corresponding
standard deviation is s, the upper error bar limit is given by ( x +s), while the lower error
bar is equal to ( x -s).

To account for variability of properties, designers use, instead of an average value of, say,
the tensile strength, the probability that the yield strength is above the minimum value
tolerable. This leads to the use of a safety factor N > 1. Thus, a working stress or safe
stress would be w = y / N.

Alternatively, a design factor (N) is defined to account for allowances thus protect
against unanticipated failure. If calculated yield stress is c, then the design stress d can
be given as d = Nc, where N is greater than unity. Thus the material to be used for a
particular application is chosen so as to have strength at least as high as the value of d.

Utilization of design stress is usually preferred since it is based on the anticipated

maximum applied stress instead of the yield strength of the material. In design it is
important to use an appropriate value for N. In general, values for factor of safety range
from 1.2 to 4.0. Higher the value of N, lesser will be the efficiency of design or material
usage i.e. either too much material or a material having a higher than necessary strength
will be used.

Selection of N will depend on a number of factors, including economics, previous

experience, the accuracy with which mechanical forces and material properties may be
determined and most importantly on the consequences of failure in terms of loss of life or
property damage. If failure would result in loss of life, the factor of safety should be
increased. In military equipment where light weight may be a prime consideration, the
factor of safety may be lower than in commercial equipment. For static loading, as in a
building, the factor of safety would be lower than in a machine, which is subjected to
vibration and fluctuating stresses.

References

1. G. E. Dieter, Mechanical Metallurgy, Third Edition, McGraw-Hill, New York,

1986.
2. M. F. Ashby and D. R. H. Jones, Engineering Materials 1, An introduction to
Their Properties and Applications, second edition, Butterworth-Heinemann,
Woburn, UK, 1996
3. William D. Callister, Jr, Materials Science and Engineering An introduction,
sixth edition, John Wiley & Sons, Inc. 2004.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 5. Diffusion

Diffusion is the process by which atoms move in a material. Many reactions in solids and
liquids are diffusion dependent. Structural control in a solid to achieve the optimum
properties is also dependent on the rate of diffusion.

Atoms are able to move throughout solids because they are not stationary but execute
rapid, small-amplitude vibrations about their equilibrium positions. Such vibrations
increase with temperature and at any temperature a very small fraction of atoms has
sufficient amplitude to move from one atomic position to an adjacent one. The fraction of
atoms possessing this amplitude increases markedly with rising temperature. In jumping
from one equilibrium position to another, an atom passes through a higher energy state
since atomic bonds are distorted and broken, and the increase in energy is supplied by
thermal vibrations. As might be expected defects, especially vacancies, are quite
instrumental in affecting the diffusion process on the type and number of defects that are
present, as well as the thermal vibrations of atoms.

Diffusion can be defined as the mass flow process in which atoms change their positions
relative to neighbors in a given phase under the influence of thermal and a gradient. The
gradient can be a compositional gradient, an electric or magnetic gradient, or stress
gradient. In this chapter we discuss diffusion because of concentration gradient only.

5.1 Diffusion mechanisms, Steady-state and Non-steady-state diffusion

5.1.1 Diffusion mechanisms

In pure metals self-diffusion occurs where there is no net mass transport, but atoms
migrate in a random manner throughout the crystal. In alloys inter-diffusion takes place
where the mass transport almost always occurs so as to minimize compositional
differences. Various atomic mechanisms for self-diffusion and inter-diffusion have been
proposed. Figure-5.1 presents schematic view of different atomic diffusion mechanisms.
The most energetically favorable process involves an interchange of places by an atom
and a neighboring vacancy vacancy diffusion. This process demands not only the
motion of vacancies, but also the presence of vacancies. The unit step in vacancy
diffusion is an atom breaks its bonds and jumps into neighboring vacant site. In
interstitial diffusion, solute atoms which are small enough to occupy interstitial sites
diffuse by jumping from one interstitial site to another. The unit step here involves jump
of the diffusing atom from one interstitial site to a neighboring site. Hydrogen, Carbon,
Nitrogen and Oxygen diffuse interstitially in most metals, and the activation energy for
diffusion is only that associated with motion since the number of occupied, adjacent
interstitial sites usually is large. Substitutional diffusion generally proceeds by the
vacancy mechanism. Thus interstitial diffusion is faster than substitutional diffusion by
the vacancy mechanism. During self-diffusion or ring mechanism or direct-exchange
mechanism, three or four atoms in the form of a ring move simultaneously round the ring,
thereby interchanging their positions. This mechanism is untenable because exceptionally
high activation energy would be required. A self-interstitial is more mobile than a
vacancy as only small activation energy is required for self-interstitial atom to move to an
equilibrium atomic position and simultaneously displace the neighboring atom into an
interstitial site. However, the equilibrium number of self-interstitial atoms present at any
temperature is negligible in comparison to the number of vacancies. This is because the
energy to form a self-interstitial is extremely large.

Figure-5.1: Diffusion mechanisms.

Diffusion in most ionic solids occurs by a vacancy mechanism. In ionic crystals, Schottky
and Frankel defects assist the diffusion process. When Frenkel defects (pair of vacancy-
interstial) dominate in an ionic crystal, the cation interstitial of the Frenkel defect carries
the diffusion flux. If Schottky defects (pair of vacant sites) dominate, the cation vacancy
carries the diffusion flux. In thermal equilibrium, in addition to above defects, ionic
crystal may have defects generated by impurities and by deviation from stochiometry.
Thus imperfections in ionic materials that influence diffusion arise in two ways: (1)
intrinsic point defects such as Frenkel and schottky defects whose number depends on
temperature, and (2) extrinsic point defects whose presence is due to impurity ions of
different valance than the host ions. The former is responsible for temperature
dependence of diffusion similar to that for self-diffusion in metals, while the latter result
in a temperature dependence of diffusion which is similar to that for interstitial solute
diffusion in metals.
For example: Cd+2 cation in NaCl crystal will results in a cation vacancy. As Schottky
defects form easily in NaCl crystal and thus cation vacancies carry the diffusion flux,
even small fraction of Cd+2 increases the diffusivity of NaCl by several orders. Excess
Zn+2 interstitials present in a non-stoichiometric ZnO compound increase the diffusivity
of Zn+2 ions significantly. It is same with non-stoichiometric FeO.

In addition to diffusion through the bulk of a solid (volume diffusion), atoms may
migrate along external or internal paths that afford lower energy barriers to motion. Thus
diffusion can occur along dislocations, grain boundaries or external surfaces. The rates of
diffusion along such short-circuit paths are significantly higher than for volume
diffusion. However, most cases of mass transport are due to volume diffusion because the
effective cross-sectional areas available for short-circuit processes are much smaller than
those for volume diffusion.

There is a difference between diffusion and net diffusion. In a homogeneous material,

atoms also diffuse but this motion is hard to detect. This is because atoms move randomly
and there will be an equal number of atoms moving in one direction than in another. In
inhomogeneous materials, the effect of diffusion is readily seen by a change in
concentration with time. In this case there is a net diffusion. Net diffusion occurs
because, although all atoms are moving randomly, there are more atoms moving away
from regions where their concentration is higher.

5.1.2 Steady-state diffusion

Diffusional processes can be either steady-state or non-steady-state. These two types of

diffusion processes are distinguished by use of a parameter called flux. It is defined as net
number of atoms crossing a unit area perpendicular to a given direction per unit time. For
steady-state diffusion, flux is constant with time, whereas for non-steady-state diffusion,
flux varies with time. A schematic view of concentration gradient with distance for both
steady-state and non-steady-state diffusion processes are shown in figure-5.2.

Figure-5.2: Steady-state and Non-steady-state diffusion processes.

Steady-state diffusion is described by Ficks first law which states that flux, J, is
proportional to the concentration gradient. The constant of proportionality is called
diffusion coefficient (diffusivity), D (cm2/sec). diffusivity is characteristic of the system
and depends on the nature of the diffusing species, the matrix in which it is diffusing, and
the temperature at which diffusion occurs. Thus under steady-state flow, the flux is
independent of time and remains the same at any cross-sectional plane along the diffusion
direction. for the one-dimensional case, Ficks first law is given by

dc 1 dn
J x = D =
dx A dt

and

J x f ( x, t )

where D is the diffusion constant, dc/dx is the gradient of the concentration c, dn/dt is the
number atoms crossing per unit time a cross-sectional plane of area A. The minus sign in
the equation means that diffusion occurs down the concentration gradient. Although, the
concentration gradient is often called the driving force for diffusion (but it is not a force
in the mechanistic sense), it is more correct to consider the reduction in total free energy
as the driving force.

An example of steady-state diffusion is provided by the permeation of hydrogen atoms

through a sheet of palladium with different imposed hydrogen gas pressures on either
side the slab. This process has been used to purify the hydrogen gas as other gases like
nitrogen, oxygen and water vapor can not diffuse through palladium.

5.1.3 Non-steady-state diffusion

Most interesting cases of diffusion are non-steady-state processes since the concentration
at a given position changes with time, and thus the flux changes with time. This is the
case when the diffusion flux depends on time, which means that a type of atoms
accumulates in a region or depleted from a region (which may cause them to accumulate
in another region). Ficks second law characterizes these processes, which is expressed
as:

dc dJ d dc
= = D
dt dx dx dx

where dc/dt is the time rate of change of concentration at a particular position, x. If D is

assumed to be a constant, then

dc d 2c
=D 2
dt dx
Solution to the above expression is possible when meaningful boundary conditions are
specified. One common set of boundary conditions can be written as

For t = 0, C = C0 at 0 x

For t > 0, C = Cs at x=0

C = C0 at x =

And the solution is

C x C0 x
= 1 erf
C s C0 2 Dt

where Cx represents the concentration at depth x after time t. The term erf stands for
Gaussian error function. Corresponding error function values for a variable are usually
found from standard mathematical tables. The above equation demonstrates the
relationship between concentration, position, and time. Thus the equation can be used to
explain many practical industrial problems like corrosion resistance of duralumin,
carburization and de-carburization of steel, doping of semi-conductors, etc.

5.2 Factors that influence diffusion and Non-equilibrium transformation &

microstructure

5.2.1 Factors that influence diffusion

Ease of a diffusion process is characterized by the parameter D, diffusivity. The value of

diffusivity for a particular system depends on many factors as many mechanisms could
be operative.

Diffusing species: If the diffusing species is able to occupy interstitial sites, then it can
easily diffuse through the parent matrix. On the other hand if the size of substitutional
species is almost equal to that of parent atomic size, substitutional diffusion would be
easier. Thus size of diffusing species will have great influence on diffusivity of the
system.

Temperature: Temperature has a most profound influence on the diffusivity and diffusion
rates. It is known that there is a barrier to diffusion created by neighboring atoms those
need to move to let the diffusing atom pass. Thus, atomic vibrations created by
temperature assist diffusion. Empirical analysis of the system resulted in an Arrhenius
type of relationship between diffusivity and temperature.

Q
D = D0 exp
RT
where D0 is a pre-exponential constant, Q is the activation energy for diffusion, R is gas
constant (Boltzmanns constant) and T is absolute temperature. From the above equation
it can be inferred that large activation energy means relatively small diffusion coefficient.
It can also be observed that there exists a linear proportional relation between (lnD) and
(1/T). Thus by plotting and considering the intercepts, values of Q and D0 can be found
experimentally.

Lattice structure: Diffusion is faster in open lattices or in open directions than in closed
directions.

Presence of defects: As mentioned in earlier section, defects like dislocations, grain

boundaries act as short-circuit paths for diffusing species, where the activation energy is
diffusion is less. Thus the presence of defects enhances the diffusivity of diffusing
species.

5.2.2 Non-equilibrium transformation & microstructure

During the processing of metallic materials, they are subjected to different conditions and
thus transformation of its structure. During casting process, liquid metal is allowed to
cool to become solid component. However, during cooling conditions can be such that
they are in equilibrium or non-equilibrium state. Phases and corresponding
microstructures, usually shown in a phase diagram, are generated during equilibrium
solidification under the conditions that are realized only for extremely slow cooling rates.
This is because with change in temperature, there must be readjustments in the
compositions of liquids and solid phases in accordance with the phase diagram. These
readjustments are accomplished by diffusional processes in both solid and liquid phases
and also across the solid-liquid interface. But, it is well understood that diffusion is time-
dependent phenomenon, and moreover diffusion in solid phases are much lower than in
liquid phases, equilibrium solidification requires extremely longer times those are
impractical. Thus virtually all practical solidification takes place under non-equilibrium
conditions, leading to compositional gradients and formation of meta-stable phases.

As a consequence of compositional gradients during non-equilibrium cooling,

segregation (concentration of particular, usually impurity elements, along places like
grain boundaries) and coring (gradual compositional changes across individual grains)
may occur. Coring is predominantly observed in alloys having a marked difference
between liquidus and solidus temperatures. It is often being removed by subsequent
annealing (incubation at relatively high temperatures that are close to lower solidus
temperature, enhances diffusion in solids) and/or hot-working, and is exploited in zone-
refining technique to produce high-purity metals. Segregation is also put to good use in
zone refining, and also in the production of rimming steel. Micro-segregation is used to
describe the differences in composition across a crystal or between neighboring crystals.
On the other hand, macro-segregation is used to describe more massive heterogeneities
which may result from entrapment of liquid pockets between growing solidifying zones.
Micro-segregation can often be removed by prolonged annealing or by hot-working; but
macro-segregation persists through normal heating and working operations.
In most situations cooling rates for equilibrium solidification are impractically slow and
unnecessary; in fact, on many occasions non-equilibrium conditions are desirable. Two
non equilibrium effects of practical importance are (1) the occurrence of phase changes
or transformations at temperatures other than those predicted by phase boundary lines on
the phase diagram, and (2) the existence of non-equilibrium phases at room temperature
that do not appear on the phase diagram. These will be detailed in next chapters about
phase transformation.

References

1. G. E. Dieter, Mechanical Metallurgy, Third Edition, McGraw-Hill, New York,

1986.
2. E. A. Brandes and G. B. Brook, Smithells Metals Reference Book, Seventh
Edition, Butterworth-Heinemann, Oxford, 1992.
3. Shewmon, P. G., Diffusion in Solids, Second Edition, The Minerals, Metals and
Materials Society, Warrendale, PA, 1989
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 7. Dislocations and Strengthening Mechanisms

6.1 Dislocations & Plastic deformation and Mechanisms of plastic deformation in

metals

6.1.1 Dislocations & Plastic deformation

While some materials are elastic in nature up point of fracture, many engineering
materials like metals and thermo-plastic polymers can undergo substantial permanent
deformation. This characteristic property of materials makes it feasible to shape them.
However, it imposes some limitations on the engineering usefulness of such materials.
Permanent deformation is due to process of shear where particles change their neighbors.
During this process inter-atomic or inter-molecular forces and structure plays important
roles, although the former are much less significant than they are in elastic behavior.
Permanent deformation is broadly two types plastic deformation and viscous flow.
Plastic deformation involves the relative sliding of atomic planes in organized manner in
crystalline solids, while the viscous flow involves the switching of neighbors with much
more freedom that does not exist in crystalline solids.

It is well known that dislocations can move under applied external stresses. Cumulative
movement of dislocations leads to the gross plastic deformation. At microscopic level,
dislocation motion involves rupture and reformation of inter-atomic bonds. The necessity
of dislocation motion for ease of plastic deformation is well explained by the discrepancy
between theoretical strength and real strength of solids, as explained in chapter-3. It has
been concluded that one-dimensional crystal defects dislocations plays an important
role in plastic deformation of crystalline solids. Their importance in plastic deformation
is relevant to their characteristic nature of motion in specific directions (slip-directions)
on specific planes (slip-planes), where edge dislocation move by slip and climb while
screw dislocation can be moved by slip and cross-slip.

The onset of plastic deformation involves start of motion of existing dislocations in real
crystal, while in perfect crystal it can be attributed to generation of dislocations and
subsequently their motion. During the motion, dislocations will tend to interact among
themselves. Dislocation interaction is very complex as number of dislocations moving on
number of slip planes in various directions. When they are in the same plane, they repel
each other if they have the same sign, and annihilate if they have opposite signs (leaving
behind a perfect crystal). In general, when dislocations are close and their strain fields
add to a larger value, they repel, because being close increases the potential energy (it
takes energy to strain a region of the material). When unlike dislocations are on closely
spaced neighboring slip planes, complete annihilation cannot occur. In this situation, they
combine to form a row of vacancies or an interstitial atom.

An important consequence interaction of dislocations that are not on parallel planes is

that they intersect each other or inhibit each others motion. Intersection of two
dislocations results in a sharp break in the dislocation line. These breaks can be of two
kinds:

(a) A jog is break in dislocation line moving it out of slip plane.

(b) A kink is break in dislocation line that remains in slip plane.

Other hindrances to dislocation motion include interstitial and substitutional atoms,

foreign particles, grain boundaries, external grain surface, and change in structure due to
phase change. Important practical consequences of hindrance of dislocation motion are
that dislocations are still movable but at higher stresses (or forces), and in most instances
that leads to generation of more dislocations. Dislocations can spawn from existing
dislocations, and from defects, grain boundaries and surface irregularities. Thus, the
number of dislocations increases dramatically during plastic deformation. As further
motion of dislocations requires increase of stress, material can be said to be strengthened
i.e. materials can be strengthened by controlling the motion of dislocation.

6.1.2 Mechanisms of plastic deformation in metals

Plastic deformation, as explained in earlier section, involves motion of dislocations.

There are two prominent mechanisms of plastic deformation, namely slip and twinning.

Slip is the prominent mechanism of plastic deformation in metals. It involves sliding of

blocks of crystal over one other along definite crystallographic planes, called slip planes.
In physical words it is analogous to a deck of cards when it is pushed from one end. Slip
occurs when shear stress applied exceeds a critical value. During slip each atom usually
moves same integral number of atomic distances along the slip plane producing a step,
but the orientation of the crystal remains the same. Steps observable under microscope as
straight lines are called slip lines.

Slip occurs most readily in specific directions (slip directions) on certain crystallographic
planes. This is due to limitations imposed by the fact that single crystal remains
homogeneous after deformation. Generally slip plane is the plane of greatest atomic
density, and the slip direction is the close packed direction within the slip plane. It turns
out that the planes of the highest atomic density are the most widely spaced planes, while
the close packed directions have the smallest translation distance. Feasible combination
of a slip plane together with a slip direction is considered as a slip system. The common
slip systems are given in table-6.1.

Table-6.1: Slip systems for different crystal structures.

Crystal Occurrence Slip planes Slip directions
FCC {111} <110>
BCC More common {110} <111>
Less common {112},{123}
HCP More common Basal plane Close packed
Less common Prismatic & Pyramidal planes directions
NaCl {110} <110>

In a single crystal, plastic deformation is accomplished by the process called slip, and
sometimes by twinning. The extent of slip depends on many factors including external
load and the corresponding value of shear stress produced by it, the geometry of crystal
structure, and the orientation of active slip planes with the direction of shearing stresses
generated. Schmid first recognized that single crystals at different orientations but of
same material require different stresses to produce slip. The dependence of various
factors has been summarized using a parameter critical resolved shear stress, R, given
as

P cos P
R = = cos cos = cos cos
A cos A
m = cos cos

where P external load applied, A cross-sectional area over which the load applied,
angle between slip direction and tensile axis, angle between normal to the slip plane
and the tensile axis and m Schmid factor.

Shear stress is maximum for the condition where = = 45 . If either of the angles are
equal to 90 , resolved shear stress will be zero, and thus no slip occurs. If the conditions
are such that either of the angles is close to 90 , crystal will tend to fracture rather than
slip. Single crystal metals and alloys are used mainly for research purpose and only in a
few cases of engineering applications.

Almost all engineering alloys are polycrystalline. Gross plastic deformation of a

polycrystalline specimen corresponds to the comparable distortion of the individual
grains by means of slip. Although some grains may be oriented favorably for slip,
yielding cannot occur unless the unfavorably oriented neighboring grains can also slip.
Thus in a polycrystalline aggregate, individual grains provide a mutual geometrical
constraint on one other, and this precludes plastic deformation at low applied stresses.
That is to initiate plastic deformation, polycrystalline metals require higher stresses than
for equivalent single crystals, where stress depends on orientation of the crystal. Much of
this increase is attributed to geometrical reasons.
Slip in polycrystalline material involves generation, movement and (re-)arrangement of
dislocations. Because of dislocation motion on different planes in various directions, they
may interact as well. This interaction can cause dislocation immobile or mobile at higher
stresses. During deformation, mechanical integrity and coherency are maintained along
the grain boundaries; that is, the grain boundaries are constrained, to some degree, in the
shape it may assume by its neighboring grains. Once the yielding has occurred, continued
plastic deformation is possible only if enough slip systems are simultaneously operative
so as to accommodate grain shape changes while maintaining grain boundary integrity.
According to von Mises criterion, a minimum of five independent slip systems must be
operative for a polycrystalline solid to exhibit ductility and maintain grain boundary
integrity. This arises from the fact that an arbitrary deformation is specified by the six
components of strain tensor, but because of requirement of constant volume, there are
only independent strain components. Crystals which do not possess five independent slip
systems are never ductile in polycrystalline form, although small plastic elongation may
be noticeable because of twinning or a favorable preferred orientation.

The second important mechanism of plastic deformation is twinning. It results when a

portion of crystal takes up an orientation that is related to the orientation of the rest of the
untwined lattice in a definite, symmetrical way. The twinned portion of the crystal is a
mirror image of the parent crystal. The plane of symmetry is called twinning plane. Each
atom in the twinned region moves by a homogeneous shear a distance proportional to its
distance from the twin plane. The lattice strains involved in twinning are small, usually in
order of fraction of inter-atomic distance, thus resulting in very small gross plastic
deformation. The important role of twinning in plastic deformation is that it causes
changes in plane orientation so that further slip can occur. If the surface is polished, the
twin would be still visible after etching because it possesses a different orientation from
the untwined region. This is in contrast with slip, where slip lines can be removed by
polishing the specimen.

Twinning also occurs in a definite direction on a specific plane for each crystal structure.
However, it is not known if there exists resolved shear stress for twinning. Twinning
generally occurs when slip is restricted, because the stress necessary for twinning is
usually higher than that for slip. Thus, some HCP metals with limited number of slip
systems may preferably twin. Also, BCC metals twin at low temperatures because slip is
difficult. Of course, twinning and slip may occur sequentially or even concurrently in
some cases. Twinning systems for some metals are given in table-6.2.

Table-6.2: Twin systems for different crystal structures.

Crystal Example Twin plane Twin direction
FCC Ag, Au, Cu (111) [112]
BCC -Fe, Ta (112) [111]
HCP Zn, Cd, Mg, Ti (1012) [1011]

Figure-6.1 presents schematic movement of atoms during plastic deformation in slip and
during twinning.
 Figure-6.1: Schematic presentation of different plastic deformation mechanism.

In table-6.3, both the mechanisms of plastic deformations are compared with respect to
their characteristics.

Table-6.3: Comparison of mechanism of plastic deformation.

during/in slip during/in twinning
Same above and below the Differ across the twin
Crystal orientation
slip plane plane
Size (in terms of
Multiples Fractions
inter-atomic distance)
Every plane of region
Occurs on Widely spread planes
involved
Time required Milli seconds Micro seconds
On many slip systems On a particular plane for
Occurrence
simultaneously each crystal

6.2 Strengthening mechanisms in Metals

Ability of a metal to deform plastically depends on ease of dislocation motion under

applied external stresses. As mentioned in earlier section, strengthening of a metal consist
hindering dislocation motion. Dislocation motion can be hindered in many ways, thus are
strengthening mechanisms in metals. Strengthening by methods of grain-size reduction,
solid-solution alloying and strain hardening applies for single-phase metals. Precipitation
hardening, dispersion hardening, fiber strengthening and Martensite strengthening are
applicable to multi-phase metallic materials.

6.2.1 Strengthening by Grain Size Reduction

This strengthening mechanism is based on the fact that crystallographic orientation

changes abruptly in passing from one grain to the next across the grain boundary. Thus it
is difficult for a dislocation moving on a common slip plane in one crystal to pass over to
a similar slip plane in another grain, especially if the orientation is very misaligned. In
addition, the crystals are separated by a thin non-crystalline region, which is the
characteristic structure of a large angle grain boundary. Atomic disorder at the boundary
causes discontinuity in slip planes. Hence dislocations are stopped by a grain boundary
and pile up against it. The smaller the grain size, the more frequent is the pile up of
dislocations. A twin boundary can also act as an obstacle to dislocation motion.

A grain boundary can hinder the dislocation motion in two ways: (1) by forcing the
dislocation to change its direction of motion and (2) discontinuity of slip plane because of
disorder. Effectiveness of grain boundary depends on its characteristic misalignment,
represented by an angle. The ordinary high-angle grain boundary (misalignment > 5 )
represents a region of random misfit between the grains on each side of the boundary.
This structure contains grain-boundary dislocations which are immobile. However they
group together within the boundary to form a step or grain boundary ledge. These ledges
can act as effective sources of dislocations as the stress at end of slip plane may trigger
new dislocations in adjacent grains. Small angle grain boundaries (misalignment < 1 )
are considered to be composed of a regular array of dislocations, and are not effective in
blocking dislocations.

With decrease in grain size, the mean distance of a dislocation can travel decreases, and
soon starts pile up of dislocations at grain boundaries. This leads to increase in yield
strength of the material. E.O.Hall and N.J.Petch have derived the following relation,
famously known as Hall-Petch relation, between yield strength (y) and grain size (d):

y = i + kd 1 2

where i is the friction stress, representing the overall resistance of the crystal lattice to
dislocation movement, k is the locking parameter that measures the relative hardening
contribution of the grain boundaries and d is the average grain diameter. Friction stress is
interpreted as the stress needed to move unlocked dislocations along the slip plane. It
depends strongly on temperature, strain, alloy and impurity content. Locking parameter is
known to be independent of temperature. Thus friction stress and locking parameters are
constants for particular material.

It is important to note that the above relation is not valid for both very large grain and
extremely fine grain sizes. Grain size reduction improves not only strength, but also the
toughness of many alloys. Grain size can be controlled by rate of cooling, and also by
plastic deformation followed by appropriate heat treatment.

Grain size is usually measured using a light microscope to observe a polished specimen
by counting the number of grains within a given area, by determining the number of
grains that intersect a given length of random line, or by comparing with standard-grain-
size charts. If d is average grain diameter, Sv is grain boundary area per unit volume, NL is
mean number of intercepts of grain boundaries per unit length of test line, NA is number
of grains per unit area on a polished surface; the all these are related as follows:

3 3 6
Sv = 2N L , d = = and d =
Sv 2N L N A
Another common method of measuring the grain size is by comparing the grains at a
fixed magnification with standard grain size charts. Charts are coded with ASTM grain
size number, G, and is related with na number of grains per mm2 at 1X magnification as

G = 2.9542 + 1.4427 ln n a

G represents number of grains per square inch (645 mm2) at a magnification of 100X is
equal to 2G-1. Higher the ASTM grain number, smaller is the grain diameter. Grain
diameter, D (in mm), and ASTM number, G, can be related as follows:

1 645
D=
100 2 G 1

6.2.2 Solid Solution Strengthening

Adding atoms of another element that those occupy interstitial or substitutional positions
in parent lattice increases the strength of parent material. This is because stress fields
generated around the solute atoms interact with the stress fields of a moving dislocation,
thereby increasing the stress required for plastic deformation i.e. the impurity atoms
cause lattice strain which can "anchor" dislocations. This occurs when the strain caused
by the alloying element compensates that of the dislocation, thus achieving a state of low
potential energy. Since solid-solution alloy additions affect the entire stress-strain curve,
it can be said that solute atoms have more influence on the frictional resistance to
dislocation motion than on the static locking of dislocations. Pure metals are almost
always softer than their alloys. Solute strengthening effectiveness depends on two factors
size difference between solute and parent atoms, and concentration of solute atoms.

Solute atoms are two categories with respect to their relative strengthening effect (1)
those produce non-spherical distortions, such as most interstitial atoms, have a relative
strengthening effect per unit concentration of about three times their shear modulus, (2)
those produce spherical distortion, such as substitutional atoms, have a relative
strengthening of about G/10.

Solute atoms interact with dislocations in many ways, namely: elastic interaction;
modulus interaction; stacking-fault interaction; electrical interaction; short-range order
interaction; and long-range order interaction. Elastic, modulus, and long-range order
interactions are of long-range i.e. they are relatively insensitive to temperature and
continue to act about 0.6 Tm where Tm is the melting temperature in absolute Kelvin
degrees.

Elastic interaction results from mutual interaction of elastic stress fields, while modulus
interaction occurs if the presence of a solute atom locally alters the modulus of the
crystal. Stacking-fault interactions arise because solute atoms may segregate to the
stacking-faults, thus lowering stacking-fault energy and widening partial dislocations.
This interaction is also called as Suzuki or Chemical interaction. Electrical interaction
arises if solute atoms of dissimilar valence interact with dislocations which have
electrical dipoles. Short-range order interaction arises from the tendency for solute atoms
to arrange themselves so that they have more then the equilibrium number of dissimilar
neighbors. The opposite of short-range order is clustering. Long-range order interaction
arises in alloys which form super-lattices, in which long-range periodic arrangement of
dissimilar atoms gets disturbed, to form anti-phase boundaries, because of dislocation
motion which leads to dissociation of dislocation into pairs of ordinary dislocations.

Some polycrystalline metals, such as mild steel, display a discrete yield point type of
behavior where a higher stress is necessary to initiate plastic flow than to continue it.
Thus, there exists a localized, heterogeneous type of transition from elastic to plastic
deformation which produces a yield point in the stress-strain curve i.e. elastic-plastic
transition is very well demarked and occurs abruptly in what is called yield-point
phenomenon. During loading the load increases steadily with elastic strain, drops
suddenly at the upper yield point where plastic deformation gets initiated. Continued
deformation fluctuates slightly about some constant stress value lower yield point.
Subsequently stress increase with increasing strain. Elongation that occurs at constant
load is called the yield-point elongation. Yield strength for metals with this phenomenon
is taken as average of lower yield point, thus it is not necessary to employ strain offset
method. A schematic picture in figure-6.2 presents the exaggerated version of yield-point
phenomenon.

Figure-6.2: Yield point phenomenon.

At the upper yield point, deformed part of metal forms a discrete band usually visible to
eye at a stress concentration. Many bands may form at the same time at different stress
concentration sites. Each band then propagates along the length of the specimen, causing
yield-point elongation. These bands are called Lders bands / Hartmann lines / stretcher
stains, and generally are approximately 45 to the tensile axis. Occurrence of yield point is
associated with presence of small amounts of interstitial or substitutional impurities. Its
been found that either unlocking of dislocations by a high stress for the case of strong
pinning or generation of new dislocations are the reasons for yield-point phenomenon.
Thus when the dislocation line is pulled free from the influence of solute atoms, slip can
occur at a lower stress. On the other hand, when dislocations are strongly pinned, new
dislocations will generate followed by drop in flow stress. This is the origin of upper
yield stress. Released dislocations tend to pile-up at grain boundaries, producing stress
concentration which in addition to applied stress unlocks sources in next grain. This is the
way band propagates. Magnitude of yield-point effect will depend on energy of
interaction between solute atoms and dislocations and on the concentration of solute
atoms at the dislocations.

6.2.3 Strain Hardening

Two most important industrial processes used to harden metals or alloys are: strain
hardening and heat treatment. Strain hardening is used for hardening/strengthening
materials that are not responsive to heat treatment. The phenomenon where ductile metals
become stronger and harder when they are deformed plastically is called strain hardening
or work hardening.

Intensity of strain hardening can be gaged from the slope of the flow curve, defined by
the parameter strain hardening exponent, n. It is measure of the ability of a metal to strain
harden. For a given amount of plastic strain, higher the value of n, greater is the strain
hardening. Increasing temperature lowers the rate of strain hardening, and thus the
treatment is given, usually, at temperatures well below the melting point of the material.
Thus the treatment is also known as cold working. Most metals strain hardens at room
temperature. The consequence of strain hardening a material is improved strength and
hardness but materials ductility will be reduced.

As mentioned in earlier chapters, with plastic deformation dislocation density increases

because under applied stress dislocation sources, like Frank-Reed source, becomes active.
Thus lightly cold worked material may have a dislocation density in order of 1012 m-2,
while a very heavily cold worked material may have 1016 m-2 against 1010 m-2 for
annealed material. High density of dislocations and thus increased interaction under
applied loads during cold working is the cause for increase in strength of a material. In
general rate of strain hardening is lower for HCP metals than for cubic metals where
higher number of slip systems is active at an instant of time. Empirically shear stress ()
to move a dislocation that increases with increase in dislocation density () is given as:

=0 + A
It is convenient to express the degree of plastic deformation as percent cold work, defined
as:

A Ad
%CW = 0 x100
A0

where A0 is the original cross sectional area that experiences deformation, and Ad is the
area after deformation.

Strain hardening is used commercially to enhance the mechanical properties of metals

during fabrication procedures. In addition to mechanical properties, physical properties of
a material also changes during cold working. There is usually a small decrease in density,
an appreciable decrease in electrical conductivity, small increase in thermal coefficient of
expansion and increased chemical reactivity (decrease in corrosion resistance).

The cold worked state is a condition of higher internal energy than the un-deformed
metal. Although cold worked dislocation cell structure is mechanically stable, it is not
thermodynamically stable. With increase in temperature state becomes more unstable,
eventually reverts to strain-free condition. This process of heating to attain strain-free
condition is called annealing heat treatment where effects of strain hardening may be
removed. Annealing process can be divided into three distinct processes: recovery,
recrystallization and grain growth. All these steps of the heat treatment process are
explained in next section. It is usual industrial practice to use alternate cycles of strain
hardening and annealing to deform most metals to a very great extent.

6.2.4-5 Precipitation Hardening and Dispersion strengthening

Small second-phase particles distributed in a ductile matrix can hinder the dislocation
motion and thus increase the strength of a material. Second-phase particles either can be
introduced by mixing and consolidation (dispersion strengthening) or precipitated in solid
state (precipitation hardening).

Precipitation hardening or age hardening is produced by solution treating and quenching

an alloy. Term Age hardening is used to describe the process because strength develops
with time. Requisite for precipitation hardening to take place is that second phase must be
soluble at an elevated temperature but precipitates upon quenching and aging at a lower
temperature. This limits the alloy systems which can be strengthened by precipitation
hardening. For example: Al-alloys, Cu-Be alloys, Mg-Al alloys, Cu-Sn alloys. If the
precipitation occurs at normal ambient temperatures, it is called natural aging. Some alloy
systems needed to be aged at higher temperatures and the process is known as artificial
aging. Most precipitation hardened alloys are limited in their maximum service
temperatures, which may lose their strength at elevated temperatures due to over-aging.

In dispersion strengthening, hard particles are mixed with matrix powder and
consolidated and processed by powder metallurgy techniques. Here second phase shall
have very little solubility in the matrix, even at elevated temperatures. Because there is
very little solubility, the particles resist growth or over-aging to a much greater extent
than the second phase particles in a precipitation hardening system. Theoretically, at
least, it is possible to produce infinite number of dispersion-hardened systems by mixing
finely divided metallic powders and second phase particles (oxides, carbides, nitrides,
borides, etc).

Dislocations moving through the matrix, which is either precipitation hardened or

dispersion strengthened, have two alternatives. They can either cut through the precipitate
particles or bend around and bypass them. The first alternative is possible only when the
slip plane is continuous from the matrix through the precipitate particle and when the
stress to move a dislocation in precipitate is comparable to that in matrix. Cutting of
particles is easier for small particles which can be considered as segregated solute atoms.
There are many particle properties that can dictate the ease of shearing, for example:
coherency strains, stacking-fault energy, ordered structure, modulus effect, interfacial
energy, morphology and lattice friction stress.

Cutting of particles is not possible when there is an interface or an abrupt change in

orientation. Under such instances, dislocations have to bend around them, and bypass.
The mechanism involved in this instance is similar to the operation of a Frank-Reed
source. Stress required to bend a dislocation is inversely proportional to the average
interspacing () of particles. Effective strengthening is achieved in the bending process,
when the particles are submicroscopic in size.

= Gb

The degree of strengthening from second phase particles depends on particle distribution
in the matrix. Particle dispersion along with the shape can be expressed by specifying the
volume fraction, average particle diameter, and mean interspacing of particles which are
interrelated. A simple expression for interspacing of particles is:

4(1 f )r
=
3f

where f is the volume fraction of spherical particles of radius r.

The spacing between second phase particles should be typically a few hundred
angstroms. Optimum strengthening occurs during aging once the right interspacing of
particles is achieved. Smaller the particles, dislocations can cut through them at lower
stresses. On the other hand, larger the particles they will be distributed at wider distances.
Thus it is necessary to arrest the growth of particles in precipitation hardened alloys at
right interspacing. Over-aging allows the fine particles to coalesce into larger and fewer
particles. The interspacing is increased thereby and the yield stress decreases.

6.2.6 Fiber strengthening

Second phase material can also be introduced into matrix in form of fibers to strengthen
it. However, mechanism of strengthening is different from either precipitation hardening
or dispersion strengthening where second phase is introduced as fine particles.

Prerequisites are materials to be used as fibers include high strength and/or high strength-
to-weight ratio. Fibers usually, thus, have high strength and high modulus while the
matrix must be ductile and non-reactive with the fibers. Fibers may be long and
continuous or they may be discontinuous. Examples for fiber material: Al2O3, boron,
graphite, metal, glass, etc. Examples for matrix material: metals, polymers. Fiber
reinforced materials are an important group of materials known as composite materials.

In fiber-reinforced materials, high modulus fibers carry essentially the entire load while
the matrix serves to transmit the load to the fibers. Matrix also protects fibers from
surface damage, serves to blunt cracks which arise from fiber breakage while it also
serves to separate the fibers. It is understood that analysis of precipitation/dispersion
strengthening is based on dislocation theory. On the other hand, the analysis of strength
of fiber-reinforced material involves the direct application of continuum principles at the
microscopic level as the material behavior is essentially elastic.

To achieve any benefit from presence of fibers, empirically it can be shown that there is a
need for critical fiber volume which must be exceeded for fiber strengthening to occur.

mu m'
f critical =
fu m'

where mu strength of strain hardened matrix, m flow stress of matrix at a strain

equal to fiber breaking stress, fu ultimate tensile strength of the fiber.

However for small values of critical fiber volume, number of fibers present may not be
sufficient to effectively restrain the elongation of the matrix. Thus a minimum volume
fraction of fiber which must be exceeded to have real reinforcement is defined as follows:

mu m'
f min =
fu + mu m'

When fiber-reinforced material is loaded, load transmits from matrix to fibers via
interface. Thus interface between matrix and fiber play an important role in fiber
strengthening.

Usually fibers are arranged in uni-directional manner, thus it is highly anisotropic in

nature. One of the consequences of the anisotropy of fiber composites is that they exhibit
shear coupling. This means that an axial load produces shear stresses, and shear load
produces axial strains. In an isotropic material a uni-axially applied load produces only
axial and transverse normal strains. Shear strains in fiber reinforced composites are
compensated practically by using a cross-ply laminate.
6.2.7 Martensite Strengthening

This strengthening can be achieved in systems where a diffusion-controlled invariant

transformation can be suppressed by rapid cooling. The Martensite strengthening process,
thus, basically is a diffusion-less and displacive reaction. The martensitic phase is formed
from the retained high temperature phase at temperatures lower then the equilibrium
invariant transformation temperature. It occurs by a process of lattices shearing.
Martensite under microscope appears as lenticular plates which divide and subdivide the
grains of the parent phase. Always touching but never crossing one another. The
characteristic lenticular shape minimizes the elastic distortion in the matrix. These
platelets grow at about one-third the velocity of sound. This high speed means activation
energy for growth is very low, and thus activation energy for nucleation determines the
amount of Martensite that can form under given conditions.

Martensite platelets attain their shape by two successive shear displacements contained in
boundaries coherent with the parent phase. The first displacement is a homogeneous
shear throughout the plate which occurs parallel to a specific plane in the parent phase
known as the habit plane. The second displacement, the lesser of the two, can take place
by one of two mechanisms: slip as in Fe-C Martensite or twinning as in Fe-Ni Martensite.
Martensite formation occurs in many systems like Fe-C, Fe-Ni, Fe-Ni-C, Cu-Zn, Au-Cd,
and even in pure metals like Li, Zr and Co. However, only the alloys based on Fe and C
show a pronounced strengthening effect.

High strength of Martensite is attributed to its typical crystal structure i.e. effective
barriers to slip are provided by the fine twin structure or the high dislocation density. In
Fe-C system, carbon atoms are also involved in strengthening. Super saturated carbon
atoms strain the ferrite lattice, and this strain can be relieved by redistribution of carbon
atoms by diffusion at room temperature. One result is that a strong binding is set up
between dislocations and the carbon atoms. This hinders the motion of dislocations, thus
increasing the strength.

This strengthening mechanism is one of the most common processes used in engineering
materials. The name basically arises from the phase Martensite that forms in steels when
they are quenched from the solutionizing temperature.

6.3 Recovery, Recrystallization and Grain Growth

As mentioned in earlier sections, annealing is an important industrial process to relieve

the stresses from cold working. During cold working grain shape changes, while material
strain hardens because of increase in dislocation density. Between 1-10% of the energy of
plastic deformation is stored in material in the form of strain energy associated with point
defects and dislocations. On annealing i.e. on heating the deformed material to higher
temperatures and holding, material tends to lose the extra strain energy and revert to the
original condition before deformation by the processes of recovery and recrystallization.
Grain growth may follow these in some instances.
6.3.1 Recovery

This is the first stage of restoration after cold working where physical properties of the
cold-worked material are restored without any observable change in microstructure. The
properties that are mostly affected by recovery ate those sensitive to point defects, for
example thermal and electrical conductivities. During recovery, which takes place at
low temperatures of annealing, some of the stored internal energy is relieved by virtue of
dislocation motion as a result of enhanced atomic diffusion. There is some reduction,
though not substantial, in dislocation density as well apart from formation of dislocation
configurations with low strain energies. Excess point defects that are created during
deformation are annihilated either by absorption at grain boundaries or dislocation
climbing process. Stored energy of cold work is the driving force for recovery.

6.3.2 Recrystallization

This stage of annealing follows after recovery stage. Here also driving force is stored
energy of cold work. Even after complete recovery, the grains are still in relatively high
strain energy state. This stage, thus, involves replacement of cold-worked structure by a
new set of strain-free, approximately equi-axed grains i.e. it is the process of nucleation
and growth of new, strain-free crystals to replace all the deformed crystals. It starts on
heating to temperatures in the range of 0.3-0.5 Tm, which is above the recovery stage.
There is no crystal structure change during recrystallization. This process is characterized
by recrystallization temperature which is defined as the temperature at which 50% of
material recrystallizes in one hour time. The recrystallization temperature is strongly
dependent on the purity of a material. Pure materials may recrystallizes around 0.3 Tm,
while impure materials may recrystallizes around 0.5-0.7 Tm.

There are many variables that influence recrystallization behavior, namely amount of
prior deformation, temperature, time, initial grain size, composition and amount of
recovery prior to the start of the recrystallization. This dependence leads to following
empirical laws:

- A minimum amount of deformation is needed to cause recrystallization.

- Smaller the degree of deformation, higher will be the recrystallization
temperature.
- The finer is the initial grain size; lower will be the recrystallization temperature.
- The larger the initial grain size, the greater degree of deformation is required to
produce an equivalent recrystallization temperature.
- Greater the degree of deformation and lower the annealing temperature, the
smaller will be the recrystallized grain size.
- The higher is the temperature of cold working, the less is the strain energy stored
and thus recrystallization temperature is correspondingly higher.
- The recrystallization rate increases exponentially with temperature.
During recrystallization, the mechanical properties that were changes during deformation
are restored to their pre-cold-work values. Thus material becomes softer, weaker and
ductile. During this stage of annealing impurity atoms tend to segregate at grain
boundaries, and retard their motion and obstruct the processes of nucleation and growth.
This solute drag effect can be used to retain cold worked strength at higher service
temperatures. Presence of second phase particles causes slowing down of
recrystallization pinning action of the particles.

6.3.3 Grain growth

This stage follows complete crystallization if the material is left at elevated temperatures.
However, grain growth does not need to be preceded by recovery and recrystallization; it
may occur in all polycrystalline materials. During this stage newly formed strain-free
grains tend to grow in size. This grain growth occurs by the migration of grain
boundaries. Driving force for this process is reduction in grain boundary energy i.e.
decreasing in free energy of the material. As the grains grow larger, the curvature of the
boundaries becomes less. This results in a tendency for larger grains to grow at the
expense of smaller grains. In practical applications, grain growth is not desirable.
Incorporation of impurity atoms and insoluble second phase particles are effective in
retarding grain growth.

Because the driving force for grain growth is lower than the driving force for
recrystallization, grain growth occurs slowly at a temperature where recrystallization
occurs at substantially high speeds. However, grain growth is strongly temperature
dependent.

References

1. William D. Callister, Jr, Materials Science and Engineering An introduction,

sixth edition, John Wiley & Sons, Inc. 2004.
2. ASM Handbook, Alloy Phase Diagrams, ASM International, Materials Park, OH,
1992.
3. ASM Handbook, Metallography and Microstructures, ASM International,
Materials Park, OH, 1985.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 6. Phase Diagrams

Many of the engineering materials possess mixtures of phases, e.g. steel, paints, and
composites. The mixture of two or more phases may permit interaction between different
phases, and results in properties usually are different from the properties of individual
phases. Different components can be combined into a single material by means of
solutions or mixtures. A solution (liquid or solid) is phase with more than one
component; a mixture is a material with more than one phase. Solute does not change the
structural pattern of the solvent, and the composition of any solution can be varied. In
mixtures, there are different phases, each with its own atomic arrangement. It is possible
to have a mixture of two different solutions!

A pure substance, under equilibrium conditions, may exist as either of a phase namely
vapor, liquid or solid, depending upon the conditions of temperature and pressure. A
phase can be defined as a homogeneous portion of a system that has uniform physical
and chemical characteristics i.e. it is a physically distinct from other phases, chemically
homogeneous and mechanically separable portion of a system. In other words, a phase is
a structurally homogeneous portion of matter. When two phases are present in a system,
it is not necessary that there be a difference in both physical and chemical properties; a
disparity in one or the other set of properties is sufficient.

There is only one vapor phase no matter how many constituents make it up. For pure
substance there is only one liquid phase, however there may be more than one solid phase
because of differences in crystal structure. A liquid solution is also a single phase, even
as a liquid mixture (e.g. oil and water) forms two phases as there is no mixing at the
molecular level. In the solid state, different chemical compositions and/or crystal
structures are possible so a solid may consist of several phases. For the same
composition, different crystal structures represent different phases. A solid solution has
atoms mixed at atomic level thus it represents a single phase. A single-phase system is
termed as homogeneous, and systems composed of two or more phases are termed as
mixtures or heterogeneous. Most of the alloy systems and composites are heterogeneous.

It is important to understand the existence of phases under various practical conditions

which may dictate the microstructure of an alloy, thus the mechanical properties and
usefulness of it. Phase diagrams provide a convenient way of representing which state of
aggregation (phase or phases) is stable for a particular set of conditions. In addition,
phase diagrams provide valuable information about melting, casting, crystallization, and
other phenomena.

Useful terminology:-

Component is either pure metal and/or compounds of which an alloy is composed. The
components of a system may be elements, ions or compounds. They refer to the
independent chemical species that comprise the system.

System it can either refer to a specific body of material under consideration or it may
relate to the series of possible alloys consisting of the same components but without
regard to alloy composition.

Solid solution it consists of atoms of at least two different types where solute atoms
occupy either substitutional or interstitial positions in the solvent lattice and the crystal
structure of the solvent is maintained.

Solubility limit for almost all alloy systems, at a specific temperature, a maximum of
solute atoms can dissolve in solvent phase to form a solid solution. The limit is known as
solubility limit. In general, solubility limit changes with temperature. If solute available is
more than the solubility limit that may lead to formation of different phase, either a solid
solution or compound.

Phase equilibrium it refers to the set of conditions where more than one phase may
exist. It can be reflected by constancy with time in the phase characteristics of a system.
In most metallurgical and materials systems, phase equilibrium involves just solid phases.
However the state of equilibrium is never completely achieved because of very slow rate
of approach of equilibrium in solid systems. This leads to non-equilibrium or meta-stable
state, which may persist indefinitely and of course, has more practical significance than
equilibrium phases. An equilibrium state of solid system can be reflected in terms of
characteristics of the microstructure, phases present and their compositions, relative
phase amounts and their spatial arrangement or distribution.

Variables of a system these include two external variables namely temperature and
pressure along with internal variable such as composition (C) and number of phases (P).
Number of independent variables among these gives the degrees of freedom (F) or
variance. All these are related for a chosen system as follows:

P+F =C+2

which is known as Gibbs Phase rule. The degrees of freedom cannot be less than zero so
that we have an upper limit to the number of phases that can exist in equilibrium for a
given system. For practical purpose, in metallurgical and materials field, pressure can be
considered as a constant, and thus the condensed phase rule is given as follows:
P + F = C +1

7.1 Equilibrium Phase Diagrams, Particle strengthening by precipitation and

precipitation reactions

7.1.1 Equilibrium Phase Diagrams

A diagram that depicts existence of different phases of a system under equilibrium is

termed as phase diagram. It is also known as equilibrium or constitutional diagram.
Equilibrium phase diagrams represent the relationships between temperature and the
compositions and the quantities of phases at equilibrium. In general practice it is
sufficient to consider only solid and liquid phases, thus pressure is assumed to be
constant (1 atm.) in most applications. These diagrams do not indicate the dynamics
when one phase transforms into another. However, it depicts information related to
microstructure and phase structure of a particular system in a convenient and concise
manner. Important information, useful for the scientists and engineers who are involved
with materials development, selection, and application in product design, obtainable from
a phase diagram can be summarized as follows:

- To show phases are present at different compositions and temperatures under

slow cooling (equilibrium) conditions.
- To indicate equilibrium solid solubility of one element/compound in another.
- To indicate temperature at which an alloy starts to solidify and the range of
solidification.
- To indicate the temperature at which different phases start to melt.
- Amount of each phase in a two-phase mixture can be obtained.

A phase diagram is actually a collection of solubility limit curves. The phase fields in
equilibrium diagrams depend on the particular systems being depicted. Set of solubility
curves that represents locus of temperatures above which all compositions are liquid are
called liquidus, while solidus represents set of solubility curves that denotes the locus of
temperatures below which all compositions are solid. Every phase diagram for two or
more components must show a liquidus and a solidus, and an intervening freezing range,
except for pure system, as melting of a phase occurs over a range of temperature.
Whether the components are metals or nonmetals, there are certain locations on the phase
diagram where the liquidus and solidus meet. For a pure component, a contact point lies
at the edge of the diagram. The liquidus and solidus also meet at the other invariant
positions on the diagram. Each invariant point represents an invariant reaction that can
occur only under a particular set of conditions between particular phases, so is the name
for it!

Phase diagrams are classified based on the number of components in the system. Single
component systems have unary diagrams, two-component systems have binary diagrams,
three-component systems are represented by ternary diagrams, and so on. When more
than two components are present, phase diagrams become extremely complicated and
difficult to represent. This chapter deals mostly with binary phase diagrams.
Unary diagrams: In these systems there is no composition change (C=1), thus only
variables are temperature and pressure. Thus in region of single phase two variables
(temperature and pressure) can be varied independently. If two phases coexist then,
according to Phase rule, either temperature or pressure can be varied independently, but
not both. At triple points, three phases can coexist at a particular set of temperature and
pressure. At these points, neither temperature nor the pressure can be changed without
disrupting the equilibrium i.e. one of the phases may disappear. Figure-1 depicts phase
diagram for water.

Figure-1: Unary phase diagram for water.

Binary diagrams: These diagrams constitutes two components, e.g.: two metals (Cu and
Ni), or a metal and a compound (Fe and Fe3C), or two compounds (Al2O3 and Si2O3), etc.
In most engineering applications, as mentioned before, condensed phase rule is
applicable. It is assumed that the same is applicable for all binary diagrams, thus the
presentation of binary diagrams becomes less complicated. Thus binary diagrams are
usually drawn showing variations in temperature and composition only. It is also to be
noted that all binary systems consist only one liquid phase i.e. a component is completely
soluble in the other component when both are in liquid state.

Hence, binary systems are classified according to their solid solubility. If both the
components are completely soluble in each other, the system is called isomorphous
system. E.g.: Cu-Ni, Ag-Au, Ge-Si, Al2O3-Cr2O3. Extent solid solubility for a system of
two metallic components can be predicted based on Hume-Ruthery conditions,
summarized in the following:

- Crystal structure of each element of solid solution must be the same.

- Size of atoms of each two elements must not differ by more than 15%.
 - Elements should not form compounds with each other i.e. there should be no
appreciable difference in the electro-negativities of the two elements.
- Elements should have the same valence.

All the Hume-Rothery rules are not always applicable for all pairs of elements which
show complete solid solubility.

In systems other than isomorphous systems i.e. in case of limited solid solubility, there
exist solid state miscibility gaps; number of invariant reactions can take place;
intermediate phases may exist over a range of composition (intermediate solid solutions)
or only at relatively fixed composition (compound). These intermediate phases may
undergo polymorphic transformations, and some may melt at a fixed temperature
(congruent transformations, in which one phase changes to another of the same
composition at definite temperature). A solid solution based on a pure component and
extending to certain finite compositions into a binary phase diagram is called a terminal
solid solution, and the line representing the solubility limit of a terminal solid solution
w.r.t a two-phase solid region is called a solvus line (figure-4).

Isomorphous system: Figure-2 depicts a typical phase diagram for an isomorphous

system made of two metallic elements A and B. As cited earlier, any phase diagram can
be considered as a map. A set of coordinates a temperature and a composition is
associated with each point in the diagram. If the alloy composition and temperature
specified, then the phase diagram allows determination of the phase or phases that will
present under equilibrium conditions. There are only two phases in the phase diagram,
the liquid and the solid phases. These single-phases regions are separated by a two-phase
region where both liquid and solid co-exist. The area in the figure-2 above the line
marked liquidus (AbB) corresponds to the region of stability of the liquid phase, and the
area below the solidus line (AdB) represents the stable region for the solid phase.
 Figure-2: Phase diagram for typical isomorphous binary system.

For the interpretation of the phase diagram, lets consider the vertical line ae drawn
corresponding to composition of 50%A +50%B and assume that the system is undergoing
equilibrium cooling. The point a on the line ae signifies that for that particular
temperature and composition, only liquid phase is stable. This is true up to the point b
which lies on the liquidus line, representing the starting of solidification. Completion of
solidification of the alloy is represented by the point, d. Point e corresponds to single-
phase solid region up to the room temperature. Point c lies in the two-phase region made
of both liquid and solid phases. Corresponding micro-structural changes are also shown
in figure-2. As shown in figure-2, above liquidus only a liquid phase exists, and below
the solidus single solid phase exists as completely solidified grains. Between these two
lines, system consist both solid crystals spread in liquid phase. It is customary to use L to
represent liquid phase(s) and Greek alphabets (, , ) for representing solid phases.

Between two extremes of the horizontal axis of the diagram, cooling curves for different
alloys are shown in figure-3 as a function of time and temperature. Cooling curves shown
in figure-3 represent A, U, X, V and B correspondingly in figure-2. Change in slope of
the cooling curve is caused by heat of fusion. In fact these changes in slope are nothing
but points on either solidus or liquidus of a phase diagram. An experimental procedure
where repeated cooling/heating of an alloy at different compositions, and corresponding
changes in slope of cooling curves will be used to construct the phase diagram.
 Figure-3: Cooling curves for isomorphous binary system.

Another important aspect of interpreting phase diagrams along with phases present is
finding the relative amount of phases present and their individual composition.

Procedure to find equilibrium concentrations of phases:

- A tie-line or isotherm (UV) is drawn across two-phase region to intersect the

boundaries of the region.
- Perpendiculars are dropped from these intersections to the composition axis,
represented by U and V in figure-2, from which each of each phase is read. U
represents composition of liquid phase and V represents composition of solid
phase as intersection U meets liquidus line and V meets solidus line.

Procedure to find equilibrium relative amounts of phases (lever rule):

- A tie-line is constructed across the two phase region at the temperature of the
alloy to intersect the region boundaries.
- The relative amount of a phase is computed by taking the length of tie line from
overall composition to the phase boundary for the other phase, and dividing by the
total tie-line length. From figure-2, relative amounts of liquid and solid phases is
given respectively by

cV Uc
CL = , CS = , and it is to be noted that C L + C S = 1 .
UV UV

Eutectic system: Many binary systems have components which have limited solid
solubility, e.g.: Cu-Ag, Pb-Sn. The regions of limited solid solubility at each end of a
phase diagram are called terminal solid solutions as they appear at ends of the diagram.
Many of the binary systems with limited solubility are of eutectic type, which consists of
specific alloy composition known as eutectic composition that solidifies at a lower
temperature than all other compositions. This low temperature which corresponds to the
lowest temperature at which the liquid can exist when cooled under equilibrium
conditions is known as eutectic temperature. The corresponding point on the phase
diagram is called eutectic point. When the liquid of eutectic composition is cooled, at or
below eutectic temperature this liquid transforms simultaneously into two solid phases
(two terminal solid solutions, represented by and ). This transformation is known as
eutectic reaction and is written symbolically as:

Liquid (L) solid solution-1 () + solid solution-2 ()

This eutectic reaction is called invariant reaction as it occurs under equilibrium

conditions at a specific temperature and specific composition which can not be varied.
Thus, this reaction is represented by a thermal horizontal arrest in the cooling curve of an
alloy of eutectic composition. A typical eutectic type phase diagram is shown in figure-4
along with a cooling curve.

As shown in figure-4, there exist three single phase regions, namely liquid (L), and
phases. There also exist three two phase regions: L+, L+ and +. These three two
phase regions are separated by horizontal line corresponding to the eutectic temperature.
Below the eutectic temperature, the material is fully solid for all compositions.
Compositions and relative amount of the phases can be determined using tie-lines and
lever rule. Compositions that are on left-hand-side of the eutectic composition are known
as hypo-eutectic compositions while compositions on right-hand-side of the eutectic
composition are called hyper-eutectic compositions. Development of micro-structure and
respective cooling curves for eutectic alloys are shown in figure-5, 6, 7 and 8 for
different compositions. The phase that forms during cooling but before reaching eutectic
temperature is called pro-eutectic phase.
 Figure-4: Typical phase diagram for a binary eutectic system.

In many systems, solidification in the solid + liquid region may lead to formation of
layered (cored) grains, even at very slow cooling rates. This is as a result of very slow or
no-diffusion in solid state compared with very high diffusion rates in liquids. The
composition of the liquid phase evolves by diffusion, following the equilibrium values
that can be derived from the tie-line method. However, new layers that solidify on top of
the grains have the equilibrium composition at that temperature but once they are solid
their composition does not change.

Figure-5: Cooling curve and micro-structure development for eutectic alloy that passes
mainly through terminal solid solution.
Figure-6: Cooling curve and micro-structure development for eutectic alloy that passes
through terminal solid solution without formation of eutectic solid.

Figure-7: Cooling curve and micro-structure development for eutectic alloy that passes
through hypo-eutectic region.

Figure-8: Cooling curve and micro-structure development for eutectic alloy that passes
through eutectic-point.

Invariant reactions: The eutectic reaction, in which a liquid transforms into two solid
phases, is just one of the possible three-phase invariant reactions that can occur in binary
systems those are not isomorphous. Schematically it can be shown as in figure-9. It
represents that a liquid phase, L, transforms into two different solids phases ( and )
upon cooling during the eutectic reaction.
 Figure-9: Schematic of eutectic invariant reaction.

In the solid state analog of a eutectic reaction, called a eutectoid reaction, one solid phase
having eutectoid composition transforms into two different solid phases. Another set of
invariant reactions that occur often in binary systems are - peritectic reaction where a
solid phase reacts with a liquid phase to produce a new solid phase, and in peritectoid
reaction, two solid phases react to form a new solid phase. Peritectic reaction is
commonly present as part of more-complicated binary diagrams, particularly if the
melting points of the two components are quite different. Peritectic and peritectoid
reactions do not give rise to micro-constituents as the eutectic and eutectoid reactions do.
Another invariant reaction that involves liquid phase is monotectic reaction in which a
liquid phase transforms into a solid phase and a liquid phase of different composition.
Over a certain range of compositions the two liquids are immiscible like oil and water
and so constitute individual phases, thus monotectic reaction can said to be associated
with miscibility gaps in the liquid state. Example system for monotectic reaction: Cu-Pb
at 954 C and 36%Pb. Analog to monotectic reaction in solid state is monotectoid
reaction in which a solid phase transforms to produce two solid phases of different
compositions. Another notable invariant reaction that is associated with liquid
immiscibility is syntectic reaction in which two liquid phases react to form a solid phase.
All the invariant reactions are summarized in the table-1 showing both symbolic reaction
and schematic part of phase diagram.

Table-1: Summary of invariant reactions in binary systems.

Reaction Symbolic equation Schematic presentation Example

Fe-C, 4.27% C,
Eutectic L+
1147 C
Fe-C, 0.80% C,
Eutectoid +
723 C
Fe-C, 0.16%C,
Peritectic L+
1495 C

Peritectoid +

Fe-C, 0.51%C,
Monotectic L1 L2 +
1495 C
 Monotectoid 1 2 +

Syntectic L1 + L2

Intermediate phases: An intermediate phase may occur over a composition range

(intermediate solid solution) or at a relatively fixed composition (compound) inside the
phase diagram and are separated from other two phases in a binary diagram by two phase
regions. Many phase diagrams contain intermediate phases whose occurrence cannot be
readily predicted from the nature of the pure components. Intermediate solid solutions
often have higher electrical resistivities and hardnesses than either of the two
components. Intermediate compounds form relatively at a fixed composition when there
exists a stoichiometric relationship between the components, for example: Mg2Ni and
MgNi2 in Mg-Ni system. These are called inter-metallic compounds, and differ from
other chemical compounds in that the bonding is primarily metallic rather than ionic or
covalent, as would be found with compounds in certain metal-nonmetal or ceramic
systems. Some metal-nonmetal compounds, Fe3C, are metallic in nature, whereas in
others, MgO and Mg2Si, bonding is mainly covalent. When using the lever rules, inter-
metallic compounds are treated like any other phase, except they appear not as a wide
region but as a vertical line.

Number of phase transformations may takes place for each system. Phase transformations
in which there are no compositional alternations are said to be congruent
transformations, and during incongruent transformations at least one of the phases will
experience a change in composition. Examples for (1) congruent transformations:
allotropic transformations, and melting of pure materials (2) incongruent transformations:
all invariant reactions, and also melting of alloy that belongs to an isomorphous system.
Intermediate phases are sometimes classified on the basis of whether they melt
congruently or incongruently. MgNi2, for example, melts congruently whereas Mg2Ni
melts incongruently since it undergoes peritectic decomposition.

7.1.2 Particle strengthening by precipitation and precipitation reactions

As explained in an earlier chapter (chapter-6: Dislocations and Strengthening

Mechanisms), by obstructing dislocation motion in different means, materials strength
can be increased. One of the methods that are applicable to multi-phase material is
particle strengthening in which second phase particles are introduced into the matrix by
either mixing-and-consolidation (dispersion strengthening) or precipitated in solid state
(precipitation hardening).

The object of the precipitation strengthening is to create in a heat-treated alloy a dense

and fine dispersion of precipitated particles in a matrix of deformable metal. The particles
act as obstacles to dislocation motion. In order for an alloy system to be able to
precipitation-strengthened for certain alloy compositions; there must be a terminal solid
solution which has a decreasing solid solubility as the temperature decreases. For
example: Au-Cu in which maximum solid solubility of Cu in Al is 5.65% at 548 C that
decreases with decreasing temperature.

The precipitation strengthening process involves the following three basic steps:

- Solutionizing (solution heat treatment), where the alloy is heated to a temperature

between solvus and solidus temperatures and kept there till a uniform solid-
solution structure is produced.
- Quenching, where the sample is rapidly cooled to a lower temperature (room
temperature) and the cooling medium is usually water. Alloy structure in this
stage consists of supersaturated solid solution.
- Aging is the last but critical step. During this heat treatment step finely dispersed
precipitate particle will form. Aging the alloy at room temperature is called
natural aging, whereas at elevated temperatures is called artificial aging. Most
alloys require artificial aging, and aging temperature is usually between 15-25%
of temperature difference between room temperature and solution heat treatment
temperature.

Precipitation strengthening and reactions that occur during precipitation can be best
illustrated using the Al-4%Cu (duralumin) system. Figure-10 depicts the Al-rich end of
the Al-Cu phase diagram. It can be observed that the alloy with 4%Cu exists as a single
phase -solid solution at around 550 C, and at room temperature as a mixture of (with
less than 0.5%Cu) and an inter-metallic compound, CuAl2 () with 52%Cu. On slow
cooling rejects excess Cu as precipitate particles of . These particles relatively coarse
in size and can cause only moderate strengthening effect.
 Figure-10: Aluminium rich end of Al-Cu phase diagram.

By rapidly cooling the alloy, a supersaturated solution can be obtained at room

temperature. As a function of time at room temperature, and at higher temperatures up to
200 C, the diffusion of Cu atoms may take place and the precipitate particles can form.
For this particular alloy, Al-4%Cu, five sequential structures can be identified: (a)
supersaturated solid solution , (b) GP1 zones, (c) GP2 zones ( phase), (d) phase
and (e) phase, CuAl2. Not all these phases can be produced at all aging temperatures.
GP1 and GP2 zones are produced at lower temperatures, and and phases occur at
higher temperatures. The initial stages of precipitation are the most difficult to analyze
because of the extremely small size of the particles and their relatively uniform
distribution. GP zones meant for Guinier-Preston zones which have a definite
composition and structure that is not the same as that of the final stable precipitate.
Evidently these particles are easier to nucleate than the final precipitate, as a result, form
first. Eventually they disappear as later more stable phases appear. and are meta-
stable transition precipitates with distinct crystal structure of their own, while is the
equilibrium stable precipitate of CuAl2.

GP1 zones:- These zones are created by Cu atoms segregating in , and the segregated
regions are of disk shape with thickness of 0.4-0.6 nm, and 8-10 nm in diameter and form
on the {100} cubic planes of the matrix. As Cu atoms which replace Al atoms are
smaller in diameter, matrix lattice strains tetragonally. These zones are said to be
coherent with the matrix lattice.

GP2 zones / phase:- With additional aging, ordering of larger clumps of Cu atoms on
{100} occurs. These zones have tetragonal structure which therefore introduces
coherency in the lattice with {100} planes of the matrix, accompanied by further
hardening. However, their size ranges from 1-4 nm thick and 10-100 nm in diameter as
aging proceeds.

phase:- This phase nucleates heterogeneously especially on dislocations. It has

tetragonal structure but is partially coherent with the matrix. This phase forms platelets
with thickness 10-150 nm.

phase:- With still further aging the equilibrium phase CuAl2 or is formed from the
transition lattice or directly from the matrix accompanied by a reduction in hardness. It
has a BCT (body-centered-tetragonal) structure, and is incoherent with the matrix. As
these particles are no longer coherent with the matrix, hardness is lower than at the stage
when coherent was present. Over-aging continues with the growth of these particles
controlled by diffusion. Variation of hardness with aging time is shown in figure-11.

The general sequence of precipitation in binary Al-Cu alloys can represented as:

Supersaturated GP1 zones GP2 zones ( phase) phase phase (CuAl2)

Most precipitation-hardening systems operate in a similar way, peak hardness usually
being attained in the later stages of coherency or at the onset of incoherency. It is quite
common for a coherent precipitate to form and then lose coherency when the particle
grows to a critical size. However, in some systems there is no evidence of coherency
strains, and the fine particles appear to act alone as impediments to dislocation
movements, for example systems with dispersion strengthening.

Figure-11: Correlation of structures and hardness for Al-4%Cu alloy.

7.2 Kinetics of nucleation and growth

Structural changes in metallic systems usually take place by nucleation and growth
whether it is just a phase change within one of the three states, or a simple structural
rearrangement within a single phase, or a phase transformation. An equilibrium phase
diagram presents the phases and phase changes expected under equilibrium conditions,
but it provides no information about the rates of transformation. Although changes in
pressure, composition, or temperature can cause phase transformations, it is temperature
changes that are more important. From a micro structural standpoint, the first process to
accompany a phase transformation is nucleation (i.e. the formation of very small particles
or nuclei of the product phase from the parent phase) of the new phase particles which are
capable of growing. The second stage is growth, in which the nucleated particles increase
their size. The transformation reaches completion if growth of these new phase particles
is allowed to proceed until the equilibrium fraction is attained.

Both nucleation and growth require that the accompanying free energy change be
negative. Consequently, the super-heating or super-cooling that is necessary for a phase
change is to be expected. That is a transformation cannot tale place precisely at the
equilibrium transformation temperature because at that temperature free energies of
phases are equal. In addition to temperature, two other factors that affect transformation
rate first, diffusion controlled rearrangement of atoms because of compositional and/or
crystal structural differences; second, difficulty encountered in nucleating small particles
via change in surface energy associated with the interface. Diffusion limits both the
nucleation and growth rates in many cases.
With the nucleation of new particle, new interface is created between the particle and
liquid. This interface will have positive energy that must be supplied during the
transformation process. A tiny particle has a large surface area to volume ratio and
therefore be unstable. Thus energy of the surface can effectively prevent the initial
formation of a tiny particle. A particle said to have nucleated when it becomes stable and
will not disappear due to thermal fluctuations. After a particle attained a critical size, it
can grow further with a continuous decrease in energy. The surface energy is no longer a
dominant factor in the growth process.

7.2.1 Nucleation kinetics

In homogeneous nucleation, the probability of nucleation occurring at any given site is

identical to that at any other site within the volume of the parent phase. When a pure
liquid metal is cooled below its equilibrium freezing temperature to a sufficient degree,
numerous homogeneous nuclei are created by slow-moving atoms bonding together.
Homogeneous nucleation usually requires a considerable amount of undercooling
(cooling a material below the equilibrium temperature for a given transformation without
the transformation occurring). Undercooling enhances the formation of nuclei that
eventually grow. If f is the free energy change accompanying the formation of a
spherical new phase particle,

4 3
f = r g + 4r 2
3

where r is the radius of the particle, g is the Gibbs free energy change per unit volume
and is the surface energy of the interface. As surface energy, , is always positive, and
g is negative, passes through a maximum. From calculus, critical values can be found
from the following:

2 16
r* = ; f hom
*
= 3 /( g ) 2
g 3

Particles which are smaller than the critical size are called embryos; those larger than the
critical size are called nuclei. As g becomes more negative with a lowering of the
temperature, the critical values of f and r becomes smaller as shown in figure-12. At
sufficiently low temperatures, nucleation can be triggered by a few atoms statistically
clustering as a nucleus, so a small critical radius is exceeded. With added growth, the new
phase attains stability. Of course atom movements are sluggish at low temperatures, so
growth is generally slow.
 Figure-12: Effect of temperature on free energy change and particle radius.

The greater the degree of undercooling below the equilibrium melting temperature of the
metal, the greater the change in volume free energy, however, the change in free energy
due to surface energy does not change much with temperature. Thus, the critical size of
nuclei is mainly determined by volume free energy. Near the freezing temperature,
critical nucleus size must be infinite since T approaches zero. As the amount of
undercooling increases, critical size decreases, and are related as follows:

2Tm
r* =
H f T

where Tm freezing temperature (in K), Hf latent heat of fusion, T amount of

undercooling at which nucleus is formed.

In heterogeneous nucleation, the probability of nucleation occurring at certain preferred

sites is much greater than that at other sites. During solidification, inclusions of foreign
particles (inoculants), walls of container holding the liquid provide preferred sites.
Irregularities in crystal structure such as point defects and dislocations possess strain
energy. In solid-solid transformation, foreign inclusions, grain boundaries, interfaces,
stacking faults and dislocations can act as preferred sites for nucleation as the strain
energy associated with them will be reduced. The released strain energy can reduce the
energy requirements for free energy change, f. Therefore, nucleation proceeds with a
smaller critical radius. A majority of reactions are initiated by some type of
heterogeneous nucleation which is common among the two types.

For example, consider the nucleation of from occurring on a foreign inclusion, , as

shown in figure-13. Considering the force equilibrium in surface tension terms,
 = cos +

where is the contact angle. An expression for f can be written in terms of volume
energy and surface energies as follows:

4
3
2 3 cos + cos 3
f het
*
= (2 3 cos + cos 3 ) = f hom
*

3( g ) 2 4

Figure-13: Schematic of heterogeneous nucleation.

By comparing the free energy terms for homogeneous and heterogeneous nucleation
processes for various contact conditions:

- When product particle makes only a point contact with the foreign surface, i.e. =
180 , the foreign particle does not play any role in the nucleation process
f het
*
= f hom
*

- If the product particle completely wets the foreign surface, i.e. = 0 , there is no
barrier for heterogeneous nucleation f het
*
=0
- In intermediate conditions such as where the product particle attains
1 *
hemispherical shape, = 0 f het
*
= f hom
2

The above derivations are helpful in selecting a heterogeneous nucleation agent. It shows
that a small contact angle is very helpful in heterogeneous nucleation. For a system of
interface, can be minimized by choosing such that energy of - interface is kept to
minimum. If the crystal structure two phases are similar and their lattice parameters are
nearly equal, energy of the interface between those two phases will be minimum. This
criterion is useful in selecting a agent for heterogeneous nucleation.

7.2.2 Growth kinetics

Many transformations occur as a result of continuous formation of critical nuclei in the

parent phase and the subsequent growth of the particles. Growth is the increase in size of
the particle after it has nucleated i.e. growth kinetics become important once an embryo
has exceeded the critical size and become a stable nucleus. Growth may proceed in two
radically different manners. In one type of growth, individual atoms move independently
from the parent to the product phase, thus it is diffusion controlled and is thermally
activated. In the other type of growth that occurs in solid-solid transformations many
atoms move cooperatively without thermal assistance. Growth that is diffusion controlled
is more common the other.

Growth usually occurs by the thermally activated jump of atoms from the parent phase to
the product phase. The unit step in the growth process thus consists of an atom leaving
the parent phase and jumping across the interface to join the product phase. At the
equilibrium temperature, both phases have the same free energy, hence the frequency of
jumps from parent phase to product phase will be equal to that from product phase to
parent phase i.e. the net growth rate is zero. At lower temperatures, product phase is
expected to have lower free energy, and thus a net flow of atoms from parent phase to
product phase. This net flux of atoms results in interface motion i.e. growth rate is taken
as the rate of increase of a linear dimension of a growing particle. As a function of
temperature, the growth rate first increases with increasing degree of supercooling, but
eventually slows-down as thermal energy decreases. This is same as for nucleation;
however the maximum in the growth rate usually occurs at a higher temperature than the
maximum in the nucleation rate. Figure-14(a) depicts the temperature dependence of
nucleation rate (U), growth rate (I) and overall transformation rate (dX/dt) that is a
function of both nucleation rate and growth rate i.e. dX/dt= fn (U, I). On the other-hand,
the time required for a transformation to completion has a reciprocal relationship to the
overall transformation rate. Temperature dependence of this time is shown in figure-14
(b). The C-curve shown in figure-14(b) is characteristic of all thermally activated
nucleation and growth transformations involving the transformation of a high-
temperature phase to low-temperature phase. This curve is also known as time-
temperature-transformation (TTT) curve. The nose of the C-curve corresponds to the
minimum time for a specified fraction of transformation. It is also the place where overall
transformation rate is a maximum.

(a) (b)
Figure-14: (a) Temperature dependence of rates, (b) Time dependence of transformation
as a function of temperature.

In many investigations, the fraction of transformation that has occurred is measured as a

function of time, while the temperature is maintained constant. Transformation progress
is usually ascertained by either microscopic examination or measurement of some
physical property. Transformation data are plotted as the fraction of transformed material
versus the logarithm of time, which results in characteristic S-curve (figure-15), largely
because diffusion plays such an important role in both nucleation and growth. Conditions
under which transformation may take place greatly affect the nature of the resulting
microstructure. At small degrees of supercooling, where slow nucleation and rapid
growth prevail, relatively coarse particles appear; at larger degrees of supercooling,
relatively fine particles result. The time dependence of the transformation rate is an
important consideration in the heat treatment of materials.

Figure-15: Fraction of transformation Vs the logarithm of time at constant temperature.

The other kind of growth involves congruent transformation which is considered as

diffusion-less because it takes place at a rate approaching the speed of sound. It can be
visualized as a cooperative type of process in which, without aid of thermal activation,
atoms move into new locations because of the strain energy resulting from like
movements of adjacent atoms. However the strains set-up in the parent phase may
impede the further transformation, thus a lower temperature or mechanical deformation
may be required to complete this martensitic transformation. The cooperative
displacement of atoms here resembles a shear process during which, for example, FCC
structure of Co transforms into HCP-Co or FCC-austenite into BCT-martensite. This
merely requires that atoms in the FCC-phase move a fraction of an inter-atomic distance.

Because of its crystallographic nature, a martensitic transformation only occurs in the

solid state. In addition, the crystal structure of the product phase must be easily generated
from that of the parent phase without diffusive motion of atoms. This is true for most
allotropic transformations in metals that occur at low temperatures or for high-
temperature transformations of metals brought about by a quench. Reasons for
martensitic transformation: (a) the free energy difference between the high-temperature
phase and low-temperature phases becomes increasingly negative with decreasing
temperature (b) the crystal structures of allotropes of a metal are relatively simple and
share similar features with each other. It is also said that diffusion-controlled nucleation
and growth and a martensitic change are competitive processes in many cases.
The martensitic transformation starts at a temperature designated Ms, which is generally
below the equilibrium temperature, Te. The transformation is completed at a lower
temperature, Mf. The amount of parent phase transformed into product phase depends on
temperature only, and is independent of time. Furthermore, in most cases, Ms temperature
and the fractional amount of product phase as a function of temperature are independent
of quenching rate. Consequently, Ms and Mf are presented as horizontal lines on a TTT
diagram. Catalytic effect of cold working can be used to make Ms approach Te.
Martensitic transformations in Fe-C alloys and Ti are of great technological importance.
Fe-C alloy transformations are dealt in detail in the following sections.

7.3 The iron carbon system, phase transformations

A study of iron-carbon system is useful and important in many respects. This is because
(1) steels constitute greatest amount of metallic materials used by man (2) solid state
transformations that occur in steels are varied and interesting. These are similar to those
occur in many other systems and helps explain the properties.

Iron-carbon phase diagram shown in figure-16 is not a complete diagram. Part of the
diagram after 6.67 wt% C is ignored as it has little commercial significance. The 6.67%C
represents the composition where an inter-metallic compound, cementite (Fe3C), with
solubility limits forms. In addition, phase diagram is not true equilibrium diagram
because cementite is not an equilibrium phase. However, in ordinary steels
decomposition of cementite into graphite never observed because nucleation of cementite
is much easier than that of graphite. Thus cementite can be treated as an equilibrium
phase for practical purposes.
 Figure-16: Iron Iron carbide phase diagram.

The Fe-Fe3C is characterized by five individual phases and four invariant reactions. Five
phases that exist in the diagram are: ferrite (BCC) Fe-C solid solution, -austenite
(FCC) Fe-C solid solution, -ferrite (BCC) Fe-C solid solution, Fe3C (iron carbide) or
cementite - an inter-metallic compound and liquid Fe-C solution. Four invariant reactions
that cause transformations in the system are namely eutectoid, eutectic, monotectic and
peritectic.

As depicted by left axes, pure iron upon heating exhibits two allotropic changes. One
involves ferrite of BCC crystal structure transforming to FCC austenite, -iron, at 910
C. At 1400 C, austenite changes to BCC phase known as -ferrite, which finally melts
at 1536 C.

Carbon present in solid iron as interstitial impurity, and forms solid solution with ferrites
/ austenite as depicted by three single fields represented by , and . Carbon dissolves
least in ferrite in which maximum amount of carbon soluble is 0.02% at 723 C. This
limited solubility is attributed to shape and size of interstitial position in BCC ferrite.
However, carbon present greatly influences the mechanical properties of ferrite.
ferrite can be used as magnetic material below 768 C. Solubility of carbon in -iron
reaches its maximum, 2.11%, at a temperature of 1147 C. Higher solubility of carbon in
austenite is attributed to FCC structure and corresponding interstitial sites. Phase
transformations involving austenite plays very significant role in heat treatment of
different steels. Austenite itself is non-magnetic. Carbon solubility in -ferrite is
maximum (0.1%) at 1495 C. As this ferrite exists only at elevated temperatures, it is of
no commercial importance. Cementite, Fe3C an inter-metallic compound forms when
amount of carbon present exceeds its solubility limit at respective temperatures. Out of
these four solid phases, cementite is hardest and brittle that is used in different forms to
increase the strength of steels. ferrite, on the other hand, is softest and act as matrix of
a composite material. By combining these two phases in a solution, a materials
properties can be varied over a large range.

For technological convenience, based on %C dissolved in it, a Fe-C solution is classified

as: commercial pure irons with less than 0.008%C; steels having %C between 0.008-
2.11; while cast irons have carbon in the range of 2.11%-6.67%. Thus commercial pure
iron is composed of exclusively ferrite at room temperature. Most of the steels and cast
irons contain both ferrite and cementite. However, commercial cast irons are not
simple alloys of iron and carbon as they contain large quantities of other elements such as
silicon, thus better consider them as ternary alloys. The presence of Si promotes the
formation of graphite instead of cementite. Thus cast irons may contain carbon in form of
both graphite and cementite, while steels will have carbon only in combined from as
cementite.

As shown in figure-16, and mentioned earlier, Fe-C system constitutes four invariant
reactions:

- peritectic reaction at 1495 C

and 0.16%C, -ferrite + L -iron (austenite)
- monotectic reaction 1495 C and 0.51%C, L L + -iron (austenite)
- eutectic reaction at 1147 C and 4.3 %C, L -iron + Fe3C (cementite)
[ledeburite]
- eutectoid reaction at 723 C and 0.8%C, -iron ferrite + Fe3C (cementite)
[pearlite]

Product phase of eutectic reaction is called ledeburite, while product from eutectoid
reaction is called pearlite. During cooling to room temperature, ledeburite transforms into
pearlite and cementite. At room temperature, thus after equilibrium cooling, Fe-C
diagram consists of either ferrite, pearlite and/or cementite. Pearlite is actually not a
single phase, but a micro-constituent having alternate thin layers of ferrite (~88%) and
Fe3C, cementite (~12%). Steels with less than 0.8%C (mild steels up to 0.3%C, medium
carbon steels with C between 0.3%-0.8% i.e. hypo-eutectoid Fe-C alloys) i.e. consists
pro-eutectoid ferrite in addition to pearlite, while steels with carbon higher than 0.8%
(high-carbon steels i.e. hyper-eutectoid Fe-C alloys) consists of pearlite and pro-eutectoid
cementite. Phase transformations involving austenite i.e. processes those involve
eutectoid reaction are of great importance in heat treatment of steels.

In practice, steels are almost always cooled from the austenitic region to room
temperature. During the cooling upon crossing the boundary of the single phase -iron,
first pro-eutectoid phase (either ferrite or cementite) forms up to eutectoid temperature.
With further cooling below the eutectoid temperature, remaining austenite decomposes to
eutectoid product called pearlite, mixture of thin layers of ferrite and cementite.
Though pearlite is not a phase, nevertheless, a constituent because it has a definite
appearance under the microscope and can be clearly identified in a structure composed of
several constituents. The decomposition of austenite to form pearlite occurs by nucleation
and growth. Nucleation, usually, occurs heterogeneously and rarely homogeneously at
grain boundaries. When it is not homogeneous, nucleation of pearlite occurs both at grain
boundaries and in the grains of austenite. When austenite forms pearlite at a constant
temperature, the spacing between adjacent lamellae of cementite is very nearly constant.
For a given colony of pearlite, all cementite plates have a common orientation in space,
and it is also true for the ferrite plates. Growth of pearlite colonies occurs not only by the
nucleation of additional lamellae but also through an advance at the ends of the lamellae.
Pearlite growth also involves the nucleation of new colonies at the interfaces between
established colonies and the parent austenite. The thickness ratio of the ferrite and
cementite layers in pearlite is approximately 8 to 1. However, the absolute layer thickness
depends on the temperature at which the isothermal transformation is allowed to occur.

The temperature at which austenite is transformed has a strong effect on the inter-
lamellar spacing of pearlite. The lower the reaction temperature, the smaller will be inter-
lamellar spacing. For example, pearlite spacing is in order of 10-3 mm when it formed at
700 C, while spacing is in order of 10-4 mm when formed at 600 C. The spacing of the
pearlite lamellae has a practical significance because the hardness of the resulting
structure depends upon it; the smaller the spacing, the harder the metal. The growth rate
of pearlite is also a strong function of temperature. At temperatures just below the
eutectoid, the growth rate increases rapidly with decreasing temperature, reaching a
maximum at 600 C, and then decreases again at lower temperatures.

Additions of alloying elements to Fe-C system bring changes (alternations to positions of

phase boundaries and shapes of fields) depends on that particular element and its
concentration. Almost all alloying elements causes the eutectoid concentration to
decrease, and most of the alloying elements (e.g.: Ti, Mo, Si, W, Cr) causes the eutectoid
temperature to increase while some other (e.g.: Ni, Mn) reduces the eutectoid
temperature. Thus alloying additions alters the relative amount of pearlite and pro-
eutectoid phase that form.

Fe-C alloys with more than 2.11% C are called cast irons. Phase transformations in cast
irons involve formation of pro-eutectic phase on crossing the liquidus. During the further
cooling, liquid of eutectic composition decomposes in to mixture of austenite and
cementite, known as ledeburite. On further cooling through eutectoid temperature,
austenite decomposes to pearlite. The room temperature microstructure of cast irons thus
consists of pearlite and cementite. Because of presence of cementite, which is hard, brittle
and white in color, product is called white cast iron. However, depending on cooling rate
and other alloying elements, carbon in cast iron may be present as graphite or cementite.
Gray cast iron contains graphite in form of flakes. These flakes are sharp and act as stress
risers. Brittleness arising because of flake shape can be avoided by producing graphite in
spherical nodules, as in malleable cast iron and SG (spheroidal graphite) cast iron.
Malleable cast iron is produced by heat treating white cast iron (Si < 1%) for prolonged
periods at about 900 C and then cooling it very slowly. The cementite decomposes and
temper carbon appears approximately as spherical particles. SG iron is produced by
adding inoculants to molten iron. In these Si content must be about 2.5%, and no
subsequent heat treatment is required.

7.4 Transformation rate effects and TTT diagrams, Microstructure and Property
Changes in Fe-C Alloys

Solid state transformations, which are very important in steels, are known to be
dependent on time at a particular temperature, as shown in figure-14(b). Isothermal
transformation diagram, also known as TTT diagram, measures the rate of transformation
at a constant temperature i.e. it shows time relationships for the phases during isothermal
transformation. Information regarding the time to start the transformation and the time
required to complete the transformation can be obtained from set of TTT diagrams. One
such set of diagram for reaction of austenite to pearlite in steel is shown in figure-17. The
diagram is not complete in the sense that the transformations of austenite that occur at
temperatures below about 550 C are not shown.

Figure-17: Partial TTT diagram for a eutectoid Fe-C alloy.

As mentioned in previous section, thickness of layers in pearlite depends on the

temperature at which the transformation occurred. If the transformation took place at a
temperature that is just below the eutectoid temperature, relatively thick layers of
ferrite and cementite are produced in what is called coarse pearlite. This is because of
high diffusion rates of carbon atoms. Thus with decreasing transformation temperature,
sluggish movement of carbon results in thinner layers ferrite and cementite i.e. fine
pearlite is produced.

At transformation temperatures below 550 C, austenite results in different product

known as bainite. Bainite also consists of ferrite and cementite phases i.e.
transformation is again diffusion controlled but morphologically it consists of very small
particles of cementite within or between fine ferrite plates. Bainite forms needles or
plates, depending on the temperature of the transformation; the microstructural details of
bainite are so fine that their resolution is only possible using electron microscope. It
differs from pearlite in the sense that different mechanism is involved in formation ob
bainite which does not have alternating layers of ferrite and cementite. In addition,
because of equal growth rates in all directions pearlite tends to form spherical colonies,
whereas bainite grows as plates and has a characteristic acicular (needlelike) appearance.
Upper bainite, formed at the upper end of the temperature range (550 C-350 C), is
characterized by relatively coarse, irregular shaped cementite particles in ferrite plates.
If the transformation is taking place at lower temperatures (350 C-250 C), the
ferrite plates assume a more regular needlelike shape, and the transformation product is
called lower bainite. At the same time carbide particles become smaller in size and
appear as cross-striations making an angle of about 55 to the axis of the ferrite plate.
Upper bainite has large rod-like cementite regions, whereas lower bainite has much finer
cementite particles as a result of sluggish diffusion of carbon atoms at lower
temperatures. Lower bainite is considerably harder than upper bainite. Another
characteristic of bainite is that as it has crystallographic orientation that is similar to that
found in simple ferrite nucleating from austenite, it is believed that bainite is nucleated by
the formation of ferrite. This is in contrast to pearlite which is believed to be nucleated by
formation of cementite.

Basically, bainite is a transformation product that is not as close to equilibrium as

pearlite. The most puzzling feature of the bainite reaction is its dual nature. In a number
of respects, it reveals properties that are typical of a nucleation and growth type of
transformation such as occurs in the formation pearlite and also a mixture of ferrite and
cementite though of quite different morphology (no alternate layers), but at the same time
it differs from the Martensite as bainite formation is athermal and diffusion controlled
though its microstructure is characterized by acicular (needlelike) appearance.

The time-temperature dependence of the bainite transformation can also be presented

using TTT diagram. It occurs at temperatures below those at which pearlite forms i.e. it
does not form until the transformation temperature falls below a definite temperature,
designated as Bs. Above this temperature austenite does not form bainite except under
B

external stresses. Below Bs, austenite does not transform completely to bainite. The
B

amount of bainite formed increases as the isothermal reaction temperature is lowered. By

reaching a lower limiting temperature, Bf, it is possible to transform austenite completely
B

to bainite. The Bs and Bf temperatures are equivalent to the Ms and Mf temperatures for
B B

Martensite.

In simple eutectoid steels, pearlite and bainite transformations overlap, thus transition
from the pearlite to bainite is smooth and continuous i.e. knees of individual pearlite and
bainite curves are merged together. However each of the transformations has a
characteristic C-curve, which can be distinguishable in presence of alloying elements. As
shown in complete TTT diagram for eutectoid steel in figure-18, above approximately
550 C-600 C, austenite transforms completely to pearlite. Below this range up to 450
C, both pearlite and bainite are formed. Finally, between 450 C and 210 C, the
reaction product is bainite only. Thus bainite transformation is favored at a high degree of
supercooling, and the pearlite transformation at a low degree of supercooling. In middle
region, pearlitic and bainitic transformations are competitive with each other.
 Figure-18: Complete TTT (isothermal transformation) diagram for eutectoid steel.

As explained in earlier section, martensitic transformation can dominate the proceedings

if steel is cooled rapid enough so that diffusion of carbon can be arrested. Transformation
of austenite to Martensite is diffusion-less, time independent and the extent of
transformation depends on the transformation temperature. Martensite is a meta-stable
phase and decomposes into ferrite and pearlite but this is extremely slow (and not
noticeable) at room temperature. Alloying additions retard the formation rate of pearlite
and bainite, thus rendering the martensitic transformation more competitive. Start of the
transformation is designated by Ms, while the completion is designated by Mf in a
transformation diagram. Martensite forms in steels possesses a body centered tetragonal
crystal structure with carbon atoms occupying one of the three interstitial sites available.
This is the reason for characteristic structure of steel Martensite instead of general BCC.
Tetragonal distortion caused by carbon atoms increases with increasing carbon content
and so is the hardness of Martensite. Austenite is slightly denser than Martensite, and
therefore, during the phase transformation upon quenching, there is a net volume
increase. If relatively large pieces are rapidly quenched, they may crack as a result of
internal stresses, especially when carbon content is more than about 0.5%.

Mechanically, Martensite is extremely hard, thus its applicability is limited by brittleness

associated with it. Characteristics of steel Martensite render it unusable for structural
applications in the as-quenched form. However, structure and thus the properties can be
altered by tempering, heat treatment observed below eutectoid temperature to permit
diffusion of carbon atoms for a reasonable period of time. During tempering, carbide
particles attain spherical shape and are distributed in ferrite phase structure called
spheroidite. Spheroidite is the softest yet toughest structure that steel may have. At lower
tempering temperature, a structure called tempered Martensite forms with similar
microstructure as that of spheroidite except that cementite particles are much, much
smaller. The tempering heat treatment is also applicable to pearlitic and bainitic
structures. This mainly results in improved machinability. The mechanism of tempering
appears to be first the precipitation of fine particles of hexagonal -carbide of
composition about Fe2.4C from Martensite, decreasing its tetragonality. At higher
temperatures or with increasing tempering times, precipitation of cementite begins and is
accompanied by dissolution of the unstable -carbide. Eventually the Martensite loses its
tetragonality and becomes BCC ferrite, the cementite coalesces into spheres. A schematic
of possible transformations involving austenite decomposition are shown in figure-19.

Figure-19: Possible transformation involving austenite decomposition.

Tempering of some steels may result in a reduction of toughness what is known as

temper embrittlement. This may be avoided by (1) compositional control, and/or (2)
tempering above 575 or below 375, followed by quenching to room temperature. The
effect is greatest in Martensite structures, less severe in bainitic structures and least
severe in pearlite structures. It appears to be associated with the segregation of solute
atoms to the grain boundaries lowering the boundary strength. Impurities responsible for
temper brittleness are: P, Sn, Sb and As. Si reduces the risk of embrittlement by carbide
formation. Mo has a stabilizing effect on carbides and is also used to minimize the risk of
temper brittleness in low alloy steels.

TTT diagrams are less of practical importance since an alloy has to be cooled rapidly and
then kept at a temperature to allow for respective transformation to take place. However,
most industrial heat treatments involve continuous cooling of a specimen to room
temperature. Hence, Continuous Cooling Transformation (CCT) diagrams are generally
more appropriate for engineering applications as components are cooled (air cooled,
furnace cooled, quenched etc.) from a processing temperature as this is more economic
than transferring to a separate furnace for an isothermal treatment. CCT diagrams
measure the extent of transformation as a function of time for a continuously decreasing
temperature. For continuous cooling, the time required for a reaction to begin and end is
delayed, thus the isothermal curves are shifted to longer times and lower temperatures.
Both TTT and CCT diagrams are, in a sense, phase diagrams with added parameter in
form of time. Each is experimentally determined for an alloy of specified composition.
These diagrams allow prediction of the microstructure after some time period for constant
temperature and continuous cooling heat treatments, respectively. Normally, bainite will
not form during continuous cooling because all the austenite will have transformed to
pearlite by the time the bainite transformation has become possible. Thus, as shown in
figure-20, region representing austenite-pearlite transformation terminates just below the
nose.

Figure-20: Superimposition of TTT and CCT diagrams for a eutectoid steel.

References

1. G. E. Dieter, Mechanical Metallurgy, Third Edition, McGraw-Hill, New York,

1986.
2. William D. Callister, Jr, Materials Science and Engineering An introduction,
sixth edition, John Wiley & Sons, Inc. 2004.
3. D. Hull, Introduction of Dislocations, Third Edition, Butterworth-Heinemann,
Woburn, UK, 1984
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 8. Failure

Failure can be defined, in general, as an event that does not accomplish its intended
purpose. Failure of a material component is the loss of ability to function normally.
Components of a system can fail one of many ways, for example excessive deformation,
fracture, corrosion, burning-out, degradation of specific properties (thermal, electrical, or
magnetic), etc. Failure of components, especially, structural members and machine
elements can lead to heavy loss of lives, wealth and even may jeopardize the society!
This chapter deals with the study of failures by mechanical means i.e. application
stresses.

Even though the causes of failure are known, prevention of failure is difficult to
guarantee. Causes for failure include: improper materials selection, improper processing,
inadequate design, misuse of a component, and improper maintenance. Its the engineers
responsibility to anticipate and prepare for possible failure; and in the event of failure, to
assess its cause and then take preventive measures.

Structural elements and machine elements can fail to perform their intended functions in
three general ways: excessive elastic deformation, excessive plastic deformation or
yielding, and fracture. Under the category of failure due to excessive elastic deformation,
for example: too flexible machine shaft can cause rapid wear of bearing. On the other
hand sudden buckling type of failure may occur. Failures due to excessive elastic
deformation are controlled by the modulus of elasticity, not by the strength of the
material. The most effective way to increase stiffness of a component is by tailoring the
shape or dimensions. Yielding or plastic deformation may render a component useless
after a certain limit. This failure is controlled by the yield strength of the material. At
room temperature, continued loading over the yielding point may lead to strain hardening
followed by fracture. However at elevated temperatures, failure occurs in form of time-
dependent yielding known as creep. Fracture involves complete disruption of continuity
of a component. It starts with initiation of a crack, followed by crack propagation.
Fracture of materials may occur in three ways brittle/ductile fracture, fatigue or
progressive fracture, delayed fracture. Ductile/brittle fracture occurs over short period of
time, and distinguishable. Fatigue failure is the mode in which most machine parts fail.
Fatigue, which is caused by a critical localized tensile stress, occurs in parts which are
subjected to alternating or fluctuating stress. Stress-rupture occurs when a metal has been
statically loaded at an elevated temperature for a long time, and is best example for
delayed fracture.

8.1 Fracture, Ductile and Brittle fracture

8.1.1 Fracture

Fracture is a form of failure, and is defined as the separation or fragmentation of a solid

body into two or more parts under the action of stress. Fracture that occurs over a very
short time period and under simple loading conditions (static i.e. constant or slowly
changing) is considered here. Fracture under complex condition, for example alternating
stress, is considered in later sections.

The process of fracture can be considered to be made up of two components, crack

initiation followed by crack propagation. Fractures are classified w.r.t. several
characteristics, for example, strain to fracture, crystallographic mode of fracture,
appearance of fracture, etc. Table-8.1 gives a brief summary of different fracture modes.

Table-8.1: Different fracture modes.

characteristic terms used
Strain to fracture Ductile Brittle
Crystallographic mode Shear Cleavage
Appearance Fibrous and gray Granular and bright
Crack propagation Along grain boundaries Through grains

Shear fracture, promoted by shear stresses, occurs as result of extensive slip on active
slip plane. On the other hand, cleavage fracture is controlled by tensile stresses acting
normal to cleavage plane. A shear fracture surface appears gray and fibrous, while a
cleavage fracture surface appears bright or granular. Actual fracture surfaces often appear
as mixture of fibrous and granular mode. Based on metallographic examination of
fracture surfaces of polycrystalline materials, they are classified as either transgranular or
intergranular. Transgranular fracture, as the name go by, represents crack propagation
through the grains, whereas intergranular fracture represents the crack that propagated
along the grain boundaries.

The fracture is termed ductile or brittle depending on the ability of a material to undergo
plastic deformation during the fracture. A ductile fracture is characterized by
considerable amount of plastic deformation prior to and during the crack propagation. On
the other hand, brittle fracture is characterized by micro-deformation or no gross
deformation during the crack propagation. Plastic deformation that occurs during ductile
fracture, if monitored, can be useful as warning sign to the fracture that may occur in later
stages. Thus brittle fracture shall be avoided as it may occur without warning!
Since deformation of a material depends on many conditions such as stress state, rate of
loading, ambient temperature, crystal structure; ductile and brittle are relative terms.
Thus the boundary between a ductile and brittle fracture is arbitrary and depends on the
situation being considered. A change from the ductile to brittle type of fracture is
promoted by a decrease in temperature, an increase in the rate of loading, and the
presence of complex state of stress (for example, due to a notch). Under the action of
tensile stresses, most metallic materials are ductile, whereas ceramics are mostly brittle,
while polymers may exhibit both types of fracture. Materials with BCC or HCP crystal
structure can be expected to experience brittle fracture under normal conditions, whereas
materials with FCC crystal structure are expected to experience ductile fracture.

Figure-8.1 depicts characteristic macroscopic fracture profiles. The profile shown in

figure-8.1(a) is representative of very high ductility represented by close to 100%
reduction in cross-sectional area. This kind of failure is usually called rupture. It is
observed in very soft metals such as pure gold and lead at room temperature and other
metals, polymers, glasses at elevated temperatures. Most ductile metals fracture preceded
by a moderate amount of necking, followed by formation of voids, cracks and finally
shear. This gives characteristic cup-and-cone fracture as shown by figure-8.1(b). In this
central interior region has an irregular and fibrous appearance. Figure-8.1(c) presents the
typical profile of brittle fracture which is usually transgranular. It occurs in most ceramics
and glasses at room temperature, long-chain polymers below their glass transition
temperatures, certain metals and alloys below their ductile-to-brittle transition
temperatures.

Figure-8.1: Fracture profiles.

Detailed and important information on the mechanism of fracture can be obtained from
microscopic examination of fracture surfaces. This study is known as fractography. This
study is most commonly done using SEM (scanning electron microscope). Common
microscopic modes of fracture observed include cleavage, quasi-cleavage, and dimpled
rupture. Characteristic feature of cleavage fracture is flat facets, and these exhibit river
marking caused by crack moving through the crystal along number of parallel planes
which form a series of plateaus and connecting ledges. Quasi-cleavage fracture is related
but distinct from cleavage in the sense that fracture surfaces are not true cleavage planes.
This often exhibit dimples and tear ridges around the periphery of the facets. Dimpled
rupture is characterized by cup-like depressions whose shape is dependent on stress state.
The depressions may be equi-axial, parabolic, or elliptical. This dimpled rupture
represents a ductile fracture. Table-8.2 distinguishes two common modes of fracture.

Table-8.2: Ductile Vs Brittle fracture.

Parameter Ductile fracture Brittle fracture
Strain energy required Higher Lower
Stress, during cracking Increasing Constant
Crack propagation Slow Fast
Warning sign Plastic deformation None
Deformation Extensive Little
Necking Yes No
Fractured surface Rough and dull Smooth and bright
Type of materials Most metals (not too cold) Ceramics, Glasses, Ice

8.1.2 Ductile fracture

Most often ductile fracture in tension occurs after appreciable plastic deformation. It
occurs by a slow tearing of the metal with the expenditure of considerable energy. It can
be said that ductile fracture in tension is usually preceded by a localized reduction in
cross-sectional area, called necking. Further it exhibits three stages - (1) after on set of
necking, cavities form, usually at inclusions at second-phase particles, in the necked
region because the geometrical changes induces hydrostatic tensile stresses, (2) the
cavities grow, and further growth leads to their coalesce resulting in formation of crack
that grows outward in direction perpendicular to the application of stress, (3) final failure
involves rapid crack propagation at about 45 to the tensile axis. This angle represents
the direction of maximum shear stress that causes shear slip in the final stage. During the
shear slip, crack propagates at a rapid speed around the outer perimeter of neck leaving
one surface in form of cup, and the other in form of cone. Thus it is known as cup-and-
cone fracture. In this central interior region has an irregular and fibrous appearance,
which signifies plastic deformation. Different progressive stages of ductile fracture are
shown in figure-8.2.
 Figure-8.2: Stages of ductile tensile fracture.

The voids are thought to be nucleated heterogeneously at sites where further deformation
is difficult. These preferred sites mainly consists of foreign inclusions, second-phase
particles like oxide particles, or even voids those can form at grain boundary triple points
in high-purity metals. It has been observed that concentration of nucleating sites had a
strong influence on ductile fracture as true strain to fracture decreases rapidly with
increasing volume fraction of second phase particles. In addition, particle shape also has
an important influence. When the particles are more spherical than plate-like, cracking is
more difficult and the ductility is increased. This is because dislocations can cross slip
around spherical particles with ease than around plate-like particles thus avoids buildup
of high stresses.

More details of fracture mechanism can be obtained from fractographic study of the
fracture surface. At high magnification under microscope, numerous spherical dimples
separated by thin walls are found on the ductile fractured surface. This is an indication
that surface had formed from numerous holes which were separated by thin walls until it
fractures. Dimples formed on shear lip of cup-and-cone fracture will be elongated
attaining parabolic shape which is indication that shear failure took place.

Ductile fracture is not particularly important in terms of mechanical behavior because it

usually is associated with good toughness. However, McClintocks analytical treatment
of ductile fracture resulted in the following equation, which gives strain to fracture, f,

(1 n) ln(l 0 2b 0 )
f =
[
sinh (1 n) ( a + b ) (2 3) ]
for a material with a stress-strain curve given by = K n , and a, b are stresses parallel
and perpendicular to the axis of the cylindrical holes respectively, is the true flow
stress, b0 is initial radius of cylindrical holes and l0 is the average spacing of holes. The
equation indicates that ductility decreases as void fraction increases, as the strain-
hardening exponent, n, decreases. This is the consequence of change of stress state from
uni-axial to tri-axial tension.
8.1.2 Brittle fracture

The other common mode of fracture is known as brittle fracture that takes place with
little or no preceding plastic deformation. It occurs, often at unpredictable levels of stress,
by rapid crack propagation. The direction of crack propagation is very nearly
perpendicular to the direction of applied tensile stress. This crack propagation
corresponds to successive and repeated breaking to atomic bonds along specific
crystallographic planes, and hence called cleavage fracture. This fracture is also said to
be transgranular because crack propagates through grains. Thus it has a grainy or faceted
texture. Most brittle fractures occur in a transgranular manner. However, brittle fracture
can occur in intergranular manner i.e. crack propagates along grain boundaries. This
happens only if grain boundaries contain a brittle film or if the grain-boundary region has
been embrittled by the segregation of detrimental elements.

In analogy to ductile fracture, as supported by number of detailed experiments, the brittle

fracture in metals is believed to take place in three stages - (1) plastic deformation that
causes dislocation pile-ups at obstacles, (2) micro-crack nucleation as a result of build-up
of shear stresses, (3) eventual crack propagation under applied stress aided by stored
elastic energy.

As mentioned earlier, brittle fracture occurs without any warning sign, thus it needs to be
avoided. Hence brittle fracture and its mechanism have been analyzed to a great extent
compared to ductile fracture. Brittle fracture usually occurs at stress levels well below
those predicted theoretically from the inherent strength due to atomic or molecular bonds.
This situation in some respects is analogous to the discrepancy between the theoretical
strength shear strength of perfect crystals and their observed lower yield strength values.

When a crystal is subjected to tensile force, separation between atoms will be increased.
The repulsive force decreases more rapidly with this than the attractive force, so that a net
force between atoms balances the tensile force. With further increase in tensile force,
repulsive force continues to decrease. At an instant, repulsive force becomes negligible
because of increased separation, which corresponds to theoretical cohesive strength of the
material. Assume that inter-atomic spacing of atoms in unstrained condition is a, and x is
change in mean inter-atomic distance. Strain, , is given by

x
=
a

And according to Hookes law, if E Youngs modulus

Ex
= E =
a

Cohesive force can be approximated with little or no error using a sine curve as follows
(refer to chapter-3: theoretical strength):
 2x
= th sin

where is wave length of the curve. For small values of x, sin x x, and thus

2x
= th

If Hookes law equation is substituted in the above equation,

E E
th =
2 a

In a brittle material, when fracture occurs, work expended goes into creation of two new
surfaces, each with a surface energy, . Work done per unit area of fracture surface, W, is
the area under the stress-strain curve.

2
2x th
W = th sin dx = = 2
0

12
2 E
= th =
th a

As brittle fracture involves, crack propagation, lets assume that a material has an interior
crack of length 2c (or a surface crack of length c) with radius of curvature as as shown
in figure-8.3.
 Figure-8.3: Schematic presentation of interior and surface cracks.

The maximum stress at the tip of crack is

c
12
c
12

max = 1 + 2 2

Before fracture starts, maximum stress at the crack tip shall rise to theoretical value of
cohesive strength. Once both are equal, crack propagates. The stress is then can be called
nominal fracture stress, f, is given by

12
E
f
4ac

The sharpest crack (i.e. maximum stress concentration) can be represented by = a.

Thus,

12
E
f
4c

Griffith theory: Griffith proposed that a brittle material contains number of micro-cracks
which causes stress rise in localized regions at a nominal stress which is well below the
theoretical value. When one of the cracks spreads into a brittle fracture, it produces an
increase in the surface energy of the sides of the crack. Source of the increased surface
energy is the elastic strain energy, released as crack spreads. Griffiths criteria for
propagation of crack include these terms as: a crack will propagate when the decrease in
elastic energy is at least equal to the energy required to create the new crack surface. This
criterion is useful in determining the tensile stress which will just cause a critical sized
crack to propagate as a brittle fracture.

Elastic energy stored under tensile stress will be released as crack propagates. Part of this
energy is expended in forming the surface of the crack, while the remaining is
transformed into kinetic energy. According to Griffith, such as crack will propagate and
produce brittle fracture when an incremental increase in its length doe not change the net
energy of the system. Strain energy released in a thin plate of unit thickness is given by
Inglis as follows:

2 c 2
Ue =
E

where E is Youngs modulus, and is the applied stress. At the same time, surface energy
gained by the system due to new surfaces formed as a crack, of length 2c, propagates can
be given as

U s = 4c

Griffiths criterion can be expressed as follows for incremental change in systems energy
with crack length:

12
U e U s 2 2 c 2 E
= = 4 =
c c E c

The equation gives the stress required to propagate a crack in a thin plate under plane
stress. The stress necessary to cause fracture varies inversely with length of existing
cracks, thus largest crack determines the strength of material. The sensitivity of the
fracture of solids to surface conditions has been termed Joffe effect. For a plate which is
thick compared with crack size (i.e. plane strain condition), the stress is given as

12
2 E
=
(1 )c
2

where is Poissons ratio. The Griffith theory applies only to completely brittle
materials. In crystalline materials which fracture in an apparently brittle mode, plastic
deformation usually occurs next to fracture surface. Orowan, thus, modified the Griffith
equation to make it more compatible by including plastic energy term as follows:

12 12
2 E ( + p ) Ep
=
c c
where p is the work of plastic deformation at the tip of the growing crack. The surface
energy term is neglected in the above equation since plastic work values are in order of
102-103 J m-2 as compared to surface energy values whose range is 1-2 J m-2.

8.2 Fracture mechanics

It is a relatively new section of materials study under mechanical loading conditions.

Using fracture mechanics concept it possible to determine whether a crack of given
length in a material with known toughness is dangerous at a given stress level. This
mechanics section can also provides guide lines for selection of materials and design
against fracture failures.

Orowan modified equation is further modified by Irwin to replace the hard to measure
plastic work term with other term that was directly measurable. New form of the equation

12
EG
= c
c

where Gc is the critical value of crack-extension force.

2 c
G=
E

G has units of J m-2, and is actually strain-energy release rate that is experimentally
measurable. The critical value of crack extension force, Gc, makes the crack propagate to
fracture, and is considered as one form of fracture toughness of the material. Fracture
toughness is defined as fracture resistance of a material in the presence of cracks.

Fracture occurs due to stress concentration at flaws, like surface scratches, voids, etc. If c
is the length of the crack and the radius of curvature at crack tip, the enhanced stress
(m) near the crack tip is given by:

12
c
m = 2

The above equation states that smaller the radius, higher is the stress enhancement.
Another parameter, often used to describe fracture toughness is known as critical stress
concentration factor, K, and is defined as follows for an infinitely wide plate subjected to
tensile stress perpendicular to crack faces:

K = c

This relation holds for specific conditions, and here it is assumed that the plate is of
infinite width having a through-thickness crack. It is worth noting that K has the unusual
units of MPam. It is a material property in the same sense that yield strength is a
material property. The stress intensity factor K is a convenient way of describing the
stress distribution around a flaw. For the general case the stress intensity factor is given
by

K = c

where is a parameter that depends on the specimen and crack sizes and geometries, as
well as the manner of load application. For example, for a plate of width w loaded in
tension with a centrally located crack of length 2c,

12
w c
K = c = c tan
c w

In this regard, it is important to realize that there are three basic modes of fracture, as
shown in figure-8.4. Mode-I corresponds to fracture where the crack surfaces are
displaced normal to themselves. This is a typical tensile type of fracture. In mode-II,
crack surfaces are sheared relative to each other in a direction normal to the edge of the
crack. In mode-III, shearing action is parallel to the edge of the crack. To indicate
different modes, it is normal practice to add the corresponding subscript. The above
example described is obviously of the mode-I. There are two extreme cases for mode-I
loading. With thin plate-type specimens the stress state is plane stress, while with thick
specimens there is a plane-strain condition. The plane-strain condition represents the
more severe stress state and the values of K are lower than for plane-stress specimens.
The fracture toughness measured under plane-strain conditions is obtained under
maximum constraint or material brittleness. The plane strain fracture toughness is
designated as KIc, and is a true material property.

Figure-8.4: Crack deformation modes.

While crack extension force, G, has more direct physical significance to the fracture
process, the stress intensity factor K is preferred in working with fracture mechanics
because it is more amenable to analytical determination. Both these parameters are
related as:

K 2 = GE for plane stress (e.g. thin plate)

K 2 = GE (1 2 ) for plane strain (e.g. thick plate)

Fracture toughness for mode-I, plane strain fracture toughness KIc, depends on many
factors, the most influential of which are temperature, strain rate, and microstructure. The
value of KIc decreases with increasing strain rate and decreasing temperature. Normally
values of KIc increases with reduction in grain size. In addition, an enhancement in yield
strength generally produces a corresponding decrease in KIc. The minimum thickness to
achieve plane-strain conditions and valid KIc measurements is

2
K
B = 2.5 Ic
0

where 0 is the 0.2% offset yield strength.

8.3 Impact fracture testing, Ductile-to-Brittle transition

8.3.1 Impact fracture testing

As mentioned in earlier section, three basic factors contribute to a brittle-cleavage type of

fracture. They are (1) tri-axial state of stress (2) a low temperature and (3) a high strain
rate or rapid rate of loading. A tri-axial state of stress, such as exists at a notch, and low
temperature are responsible for most service failures of brittle type. Since, these effects
are accented at a high rate of loading, various types of notched-bar impact tests have been
used to determine the susceptibility of materials to brittle fracture.

The changes produced by the introduction of a notch have important consequences in the
fracture process. The chief effect of the notch is not in introducing a stress concentration
but in producing a tri-axial state of stress at the notch. As a result of this tri-axial state of
stress, yield strength of a notched specimen is greater than the uni-axial yield strength
because it is more difficult to spread the yielded zone in the presence of tri-axial stresses.
The ratio of notched-to-unnotched strength is referred as the plastic-constraint factor, q.
Unlike stress concentration factor, which can reach values in excess of 10 as the notch is
made sharper and deeper, Orowan has shown that the plastic constraint factor cannot
exceed a value of 2.57. It results in notch-strengthening of a ductile material. But in a
material prone to brittle fracture the increased tensile stresses from the plastic constraint
can exceed the critical value for fracture before the material undergoes general plastic
yielding. A notch, thus, increases the tendency for brittle fracture in four important ways:
(a) By producing high local stresses, (b) by introducing a tri-axial state of stress, (c) by
producing high local strain hardening and cracking, and (d) by producing a local
magnification to the strain rate.

Two classes of specimens have been standardized for notch-impact testing: (i) Charpy
specimen. It has a square cross section and contains 45 V notch, 2 mm deep with a 0.25
mm root radius. The specimen is supported as a horizontal beam, and loaded behind the
notch by the impact of heavy swinging pendulum with velocity about 5 m/sec. The
energy expended is measured in terms of change in potential energy (height) of the
pendulum. The specimen is forced to bend and fracture at a high strain rate in order of
103 sec-1. The V-notch in the specimen provides triaxiality of stress, and the high hammer
velocity insures a high strain rate. (ii) Izod specimen. It has either circular or square cross
section and contains V notch near the clamped end. It is supported as over hanging
vertical beam. Load is again applied using swinging pendulum but in front of the notch.
Figure-8.5 depicts loading of Charpy and Izod specimens for notch-impact tests.

Figure-8.5: Loading of Charpy and Izod notched-bar impact testing specimens.

The principal measurement from the impact test is the energy absorbed in fracturing the
specimen. Energy expended during fracture is some times known as notch toughness. The
energy expended will be high for complete ductile fracture, while it is less for brittle
fracture. However, it is important to note that measurement of energy expended is only a
relative energy, and can not be used directly as design consideration. Another common
result from the Charpy test is by examining the fracture surface. It is useful in
determining whether the fracture is fibrous (shear fracture), granular (cleavage fracture),
or a mixture of both. A third measurement that can be made is the ductility, indicated by
the percent contraction of the specimen at the notch. The notched-bar impact test is most
useful when conducted over a range of temperatures. Thus it is possible to know at which
temperature the ductile-to-brittle transition is taking place. Thus the chief engineering use
of Charpy test is in selecting materials which are resistant to brittle fracture by means of
transition temperature curves. When selecting a material for design purposes which might
be susceptible to brittle fracture, the material with the lowest transition temperature is to
be preferred.
Both plane strain fracture toughness and notched-bar impact tests determine the fracture
properties of materials. The former are quantitative in nature, in that a specific property is
determined. On the other hand, the results of the impact tests are qualitative in nature,
thus are of little use fro design purposes. Plane strain fracture toughness tests are not as
simple to perform as impact tests, and equipment and specimens are more expensive.

8.3.2 Ductile-to-Brittle transition

It is well understood that ductile and brittle are relative, and thus interchange between
these two modes of fracture is achievable with ease. The term Ductile-to-Brittle
transition (DBT) is used in relation to the temperature dependence of the measured
impact energy absorption. For a material, as the temperature is lowered, the impact
energy drops suddenly over a relatively narrow temperature range, below which the
energy has a considerably lower value as a representative of brittle fracture.

The temperature where DBT occurs is termed as Ductile-to-Brittle Transition

Temperature (DBTT). A typical variation of energy expended as a function of
temperature is shown in figure-8.6. From the figure it can be concluded that there is no
single criterion that defines the transition temperature. Above the DBTT, the yield
strength (y) is lower than the tensile stress necessary to cause brittle failure (f) i.e. y <
f. With decreasing temperature, the yield strength increases rapidly to the point where it
equals the tensile stress for brittle failure, and below this temperature, fracture usually
occurs in brittle/cleavage mode. So, at and below the DBTT, y = f. At the transition
temperature, the micro-cracks that form are of critical size for crack propagation, and at
lower temperatures these cracks exceed the critical size. Over a temperature range just
above DBTT, micro-cracks formed are initially sub-critical so that further plastic
deformation and strain hardening must proceed before the tensile stress level becomes
sufficient to cause crack propagation. With further increase in temperature, micro-cracks
no longer form and fracture mode changes from cleavage to ductile.

Figure-8.6: Typical variation of fracture toughness as a function of temperature.

As shown in the above figure, BCC metals possess a distinct DBTT compared with other
metals. Common BCC metals are to become brittle at low temperatures or at extremely
high rates of strain. Many FCC metals, on the other hand, remain ductile even at very low
temperatures. In metals DBTT is around 0.1-0.2 Tm while in ceramics it is about 0.5-0.7
Tm, where Tm represents absolute melting temperature. The crack propagation stress
usually is relatively insensitive to temperature. Strain rate which increases the yield
strength but not the crack propagation stress increases the DBTT. As mentioned earlier,
DBTT does not have a unique value for a given material, and is a function of several
other variables such as flaw size, strain rate, triaxiality of stress, etc. Sharp notches in the
specimen provide stress concentration centers and thus increase the DBTT. Fine grained
materials have a lower transition temperature as compared to coarse grained materials.

There are other factors which affect the DBTT for a material, for example metallurgical
factors. One of the most important is microstructure, which in turn can depend on heat
treatment practice and temperature, for example, for steels. The composition also has a
very pronounced effect on DBTT. In steels, it is found that impurities like P, Si, Mo, Cr
along with C increase the DBTT, while Mn and Ni have reverse effect.

8.4 Fatigue, Crack initiation and propagation, Crack propagation rate

8.4.1 Fatigue

Failures occurring under conditions of dynamic or alternating loading are called fatigue
failures, presumably because it is generally observed that these failures occur only after a
considerable period of service. Fatigue failure usually occurs at stresses well below those
required for yielding, or in some cases above the yield strength but below the tensile
strength of the material. These failures are dangerous because they occur without any
warning. Typical machine components subjected to fatigue are automobile crank-shaft,
bridges, aircraft landing gear, etc. Fatigue failures occur in both metallic and non-metallic
materials, and are responsible for a large number fraction of identifiable service failures
of metals.

A typical fatigue-fracture surface looks like the one shown in figure-8.7. The fatigue
crack nucleates at the stress concentration. Generally, the fatigue fracture surface is
perpendicular to the direction of an applied stress. A fatigue failure can be recognized
from the appearance of the fracture surface, which shows a smooth and polished surface
that corresponds to the slow growth of crack, when the crack faces smoothen out by
constant rubbing against each other and a rough/granular region corresponds to the stage
of fast growth, after critical conditions is attained where member has failed in a ductile
manner when cross section was no longer able to carry the applied load. The region of a
fracture surface that formed during the crack propagation step may be results in
characteristic pattern of concentric rings spread over the smooth region of the fracture
surface, known as beach marks or striations, radiating outward from the point of
initiation of the failure, as shown in figure-8.7. Beach marks (also known as clamshell
pattern) are macroscopic dimensions and may be observed with the unaided eye. These
markings are found for components that experienced interruptions during the crack
propagation stage. Each beach mark band represents a period of time over which crack
growth occurred. On the other hand fatigue striations are microscopic in size and subject
to observation with the electron microscope (either TEM or SEM). The relatively widely
spaced striations are caused by variations in the stress amplitude during the life of the
component. On a much finer level, a large number of striations may be sometimes being
seen. The width of each striation here is equal to the distance by which the crack grows
during one cycle. Any point with stress concentration such as sharp corner or notch or
metallurgical inclusion can act as point of initiation of fatigue crack.

Figure-8.7: Schematic of fatigue fracture surface.

Three basic requisites for occurrence of fatigue fracture are: (a) a maximum tensile stress
of sufficiently high value (b) a large enough variation or fluctuation in the applied stress
and (c) a sufficiently large number of cycles of applied stress. The stress cycles that are
evident in fatigue studies are characterized using many parameters, such as mean stress,
alternating stress, stress ratio and amplitude ratio. If the applied stress varies between max
and min,

Range of stress, r = max min

Alternating stress, a = r/2 = (max min)/2

Mean stress, m = (max + min)/2

Stress ratio, R= min / max

Amplitude ratio, A= a / m = (1-R) / (1+R)

There are as many ways of conducting a test to measure fatigue as there are ways of
applying repeated stresses to a metal. A specimen of rotating beam type is often used
because of its simplicity. In a fatigue test, the premium output is number of cycles
required to fracture the specimen at a given stress. If the maximum tensile stress applied
is only slightly less than the yield strength (or tensile strength), test will run only a few
cycles. Continues reduction of the stress greatly increases the life of the specimen, hence
fatigue data is usually presented by plotting maximum stress (S) against number of cycles
to fracture (N), using a logarithmic scale for the latter variable. This form of curve, S-N
curve, shown in figurer-8.8, is significant because there is a stress below which the
specimens do not fracture. This limiting stress is called fatigue limit or endurance limit
(e), below which fatigue will never occur. Fatigue limit can be related to tensile strength
of the material, and the ratio of fatigue limit to tensile strength is known as endurance
ratio. Most steels have distinct fatigue limit, and is usually about 0.4-0.5 of tensile
strength i.e. endurance ratio=0.4-0.5. Unlike steels, most nonferrous metals do not have a
fatigue limit i.e. S-N curve continues to fall steadily with decreasing stress, though at a
decreasing rate. Thus, fatigue will ultimately occur regardless of the magnitude of the
applied stress. Fatigue response of these materials is specified for a number of stress
cycles, normally 107, and is known as fatigue strength. Another important parameter that
characterizes a materials fatigue behavior is fatigue life, Nf, number of cycles to cause
fatigue failure at a specified stress level.

Figure-8.8: Typical S-N curves for ferrous and non-ferrous metals.

The S-N curve in the high-cycle region is some times described by the Basquin equation:

N ap = C

where a is the alternating stress or stress amplitude and p and C are empirical constants.
It will be generally found that there is a considerable amount of scatter in the results.
Thus fatigue life and fatigue strength are considered as statistical quantities. It has been
observed that scatter in fatigue life decreases with increase in stress. The statistical
problem of accurately determining the fatigue limit is complicated by the fact that
complete S-N curve cannot obtainable using a single specimen as specimen cannot be
rested during the test.
With the exception of rotating shafts, most structural members are not subjected to
symmetrical stress cycles having a mean stress of zero. When (R = -1 for the case of
completely reversed stress) R becomes more positive i.e. with increasing mean stress, the
measured fatigue limit becomes greater (figure-8.9). For a given stress amplitude, a mean
tensile stress reduces the fatigue life of a material. It has been observed that as the mean
stress becomes more positive the allowable alternating stress decreases (figure-8.10).

Figure-8.9: Dependence of fatigue limit on stress ratio.

Figure-8.10: Dependence of alternating stress on mean stress.

The Goodman diagram presents the dependence of allowable stress ranges on mean stress
for a material. As shown in figure-8.11, allowable stress range increases with increasing
compressive mean stress i.e. compressive stress increases the fatigue limit.
 Figure-8.11: Goodman diagram.

An alternative method of presenting mean stress data is by using Heig-Soderberg diagram

in which alternating stress is plotted against the mean stress. As figure-8.12 depicts,
Goodmans criterion appears as a straight line. Test data for ductile metals, however,
follows parabolic curve proposed by Gerber. Both these criteria can be expressed as:

x

a = e 1 m
u

where x=1 for Goodman line, x=2 for the Gerber parabola, and e is the fatigue limit for
completely reversed loading. Soderberg line presents the data when the design is based
on yield strength (0).

Figure-8.12: Heig-Soderberg diagram.

8.4.2 Crack initiation and propagation

Based on structural changes that occur during fatigue, fatigue failure process can be
considered as made of following stages: (a) crack initiation includes the early
development of fatigue damage that can be removed by suitable thermal anneal (b) slip-
band crack growth involves the deepening of initial crack on planes of high shear stress.
This is also known as stage-I crack growth. (c) crack growth on planes of high tensile
stress involves growth of crack in direction normal to maximum tensile stress, called
stage-II crack growth (d) final ductile failure occurs when the crack reaches a size so
that the remaining cross-section cannot support the applied load.

Fatigue failures usually are found to initiate at a free surface or at internal flaws such as
inclusions where the local stress causes some heterogeneous permanent flow leading to
formation of a small crack. Fatigue failures start as small microscopic cracks and,
accordingly, are very sensitive to even minute stress raisers. It has been observed that
diffusion processes are not necessary to the formation of fatigue cracks. The initiation of
a fatigue crack does not lead to immediate failure, rather, the crack propagates slowly and
discontinuously across the specimen under the action of cyclic stress. The amount of
crack motion per cycle depends on the material and the stress level; high stresses favor
larger crack growth increments per cycle. Eventually, the crack propagates to the point
where the remaining intact cross section of material no longer can support the applied
load, and further crack propagation is rapid, leading to catastrophic failure. The final
fracture surface is composed of an area over which there was slow crack propagation and
an area where the crack moved rapidly. Final fracture can be either ductile or brittle type.

In polycrystalline metals, during a fatigue test slip lines appear first on crystal whose slip
planes have the highest resolved shear stress. As time goes on and the number of stress
cycles increases, the size and number of slip bands (clusters of slip lines) increase. The
extent and number of slip bands are also a function of the amplitude of the applied stress;
higher stresses give larger values. In fatigue, under cyclic loading, the slip bands tend to
group into packets or striations in a slip band. Each striation represents the successive
position of an advancing crack front that is normal to the greatest tensile stress. Ridge
kinds of striations are called extrusions while, while crevice striations are known as
intrusions, and both tend to be formed depending on the crystal orientation. It has been
shown that fatigue cracks initiate at intrusions and extrusions. These surface disturbances
are approximately 10-3 to 10-4 mm in height and appear as early as 1/10 of the total life of
a specimen. Figure-8.13 depicts the slip bands forming under static load and cyclic load.
Table-8.3 summarizes deformation features under static and cyclic loading.

With increasing numbers of cycles, the surface grooves deepen and the crevices or
intrusions take on the nature of a crack. When this happens, stage-I of the crack-growth
process has begun i.e. stage-I crack propagates along persistent slip bands, and can
continue for a large fraction of the fatigue life. Low applied stresses and deformation by
slip on a single slip plane favor stage-I growth. On the other hand multiple-slip
conditions favor stage-II growth. The transition from stage-I to stage-II is often induced
when a slip-plane crack meets an obstacle such as a grain boundary. The rate of crack
propagation in stage-I is generally very low on the order of nm per cycle compared with
that in stage-II where it is in order of m per cycle. Thus, from a practical viewpoint,
stage-II is of importance than stage-I. Stage-I growth follows a slip plane, whereas stage-
II growth does not have this crystallographic character. The fracture surface of stage-I is
practically featureless. On the other hand, stage-II shows a characteristic pattern of
striations, which occurs by repetitive plastic blunting and sharpening of the crack tip.
Table-8.4 distinguishes stage-I from stage-II crack growth.

Figure-8.13: Comparison of slip bands formed under (a) static loading and (b) cyclic
loading.

Table-8.3: Deformation under static and cyclic loads.

Feature Static load Cyclic load
Slip (nm) 1000 1-10
Deformation feature Contour Extrusions & Intrusions
Grains involved All grains Some grains
Vacancy concentration Less Very high
Necessity of diffusion Required Not necessary
 Table-8.4: Fatigue crack growth: stage-I Vs stage-II.
Stage-I Stage-II
Stresses involved Shear Tensile
Crystallographic orientation Yes No
Crack propagation rate Low (nm/cycle) High (m/cycle)
Slip on Single slip plane Multiple slip planes
Feature Feature less Striations

The region of fatigue fracture surface that formed during the crack propagation step can
be characterized by two types of markings termed beach marks and striations. Both of
these features indicate the position of the crack tip at some point in time and appear as
concentric rings that expand away from the crack initiation site(s), frequently in a circular
or semicircular pattern. Beach marks (some times also called clamshell marks) are of
macroscopic dimensions, found for components that experienced interruptions during the
crack propagation stage. Each beach mark band represents a period of time over which
crack growth occurred. Striations are microscopic in size, and each of it is thought to
represent the advance distance of a crack front during a single load cycle. Striation width
depends on, and increases with, increasing stress range. There may be literally thousands
of striations within a single beach mark. The presence of beach marks/striations on a
fracture surface confirms that the cause of failure in fatigue. Nevertheless, the absence of
either or both does not exclude fatigue as the case of failure.

8.4.3 Crack propagation rate

As pointed out in earlier section, crack propagation in stage-II, where it is faster, has
greater practical importance. Considerable research has gone into studying the crack
propagation in this stage as the results from this can be used as fail-safe design
considerations.

To distinguish the crack propagation time from crack initiation time, pre-existing flaws or
cracks are introduced into a fatigue specimen. Crack propagation rate (da/dN) is studied
in terms of increase in crack growth which is a proportional to number of applied stress
cycles and length of the crack i.e.

da
= fn( , a ) = C am a n
dN

where C a constant, a the alternating stress, a the crack length, m=2-4, and n=1-2.

It has been found that crack growth rate can be related to stress-intensity factor, K (mode-
I) of fracture mechanics which itself is a combination of stress and crack length. Thus,
da
= A(K ) p
dN

where K stress-intensity factor range., A and p constants that are functions of

material, environment, frequency, temperature and stress ratio.

K = K max K min = max a min a = r a

Since the stress intensity factor is not defined for compressive stresses, if min is in
compression, zero value is assigned for Kmin.

Fatigue data presented in terms of log (crack growth rate) against log (K) is shown in
figure-8.14. The curve can be divided into three regions. In region-I, which is bound by a
threshold value Kth, no observable crack growth occurs i.e. crack growth rate is very
slow in order of 0.25 nm per cycle or less. The straight line part of curve in region-II can
be represented by the power law (also known as Paris law)

da
= A(K ) p
dN

and slope of the straight line will give value for p. The values of p are the range of 3 for
steels, and in the range of 3-4 for aluminium alloys. Region-III of the curve represents the
rapid unstable crack growth to ultimate fracture. Here Kmax approaches Kc, the fracture
toughness of the material. When K is known under relevant loading conditions, the
fatigue life can be estimated by integrating the above equation between limits of initial
crack size and final crack size. One such equation is given below when p2:

a f ( m / 2) +1 a0 ( m / 2)+1
Nf =
A p p 2 p (( p / 2) + 1)

Nf number of cycles required for a crack to grow from a0 to af. - is a parameter that
depends on the specimen and crack sizes and geometries, as well as the manner of load
application.
 Figure-8.14: Schematic presentation of fatigue crack growth.

Increasing mean stress, R, in the fatigue cycle has a tendency to increase the crack growth
rates in all portions of the curve. The effect of increasing R is less in region-II than in
regions I and III. The influence of R on the region-II is given by:

da A(K ) p
=
dN (1 R) K c K

where Kc fracture toughness, R stress ratio. Fatigue life estimation is usually carried
out when R=-1 i.e. fully reversed stress conditions, whereas fatigue crack growth data is
determined for R=0 i.e. pulsating tension without compression. Compression loads are
avoided because during compression loading the crack is closed and the stress intensity
factor is zero.

8.5 Creep, Generalized creep behavior, Stress and Temperature effects

8.5.1 Creep

It is well known and observed number of times that shear modulus of a material increases
with decreasing temperature, and this has a corresponding effect on the flow-stress, since
the stress associated with a dislocation is always proportional to the shear modulus. Thus
/E can be expected to be constant with varying temperature. However, this ratio
decreases with increase in temperature. Since, the dislocation-density contribution of the
flow stress is effectively constant, it was concluded that there must be a second basic
component of flow stress that is temperature-dependent i.e.

= * +E

where * is thermal component and E is athermal component of flow stress. The fact that
the flow stress contains a component that responds to thermal activation implies that
plastic deformation can occur while both the temperature and the stress are maintained
constant. Deformation that occurs under these conditions which is time-dependent is
known as creep. Creep deformation (constant stress) is possible at all temperatures above
absolute zero. However, it is extremely sensitive to temperature. Hence, creep in usually
considered important at elevated temperatures (temperatures greater than 0.4 Tm, Tm is
absolute melting temperature).

8.5.2 Generalized creep behavior

Creep behavior of a material is studied using creep test. In creep test, the tensile specimen
is subjected to a constant load or stress at a constant temperature. Most creep tests are
conducted at constant load in analogous to engineering application, whereas creep tests at
constant stress are necessary for understanding of mechanism of creep. During the creep
test, strain (change in length) is measured as a function of elapsed time. Creep test data is
presented as a plot between time and strain known as creep curve. Figure-8.15 depicts a
typical creep curve. The slope of the creep curve is designated as creep rate.

Figure-8.15: Typical creep curve.

As shown in the above figure, upon loading the specimen, there is an instantaneous
deformation (0) that is mostly elastic. Actual creep curve follows this elastic
deformation. Based on the variation of creep rate with time, creep curve is considered to
be consists of three portions, each of which has its own distinctive strain-time feature.
After initial rapid elongation, 0, the creep rate decreases continuously with time, and is
known as primary or transient creep. Primary creep is followed by secondary or steady-
state or viscous creep, which is characterized by constant creep rate. This stage of creep
is often the longest duration of the three modes. Finally, a third stage of creep known as,
tertiary creep occurs that is characterized by increase in creep rate. Creep curve could be
represented by the following equation, according to Andrade:

= 0 (1 + t 1 3 )e kt

where is strain in time t, and and k are constants. A better equation is proposed by
Garofalo:

= 0 + t (1 e rt ) + & s t

where 0 instantaneous strain on loading, t limit for transient creep, r ratio of

transient creep rate to the transient creep strain, s steady-state creep rate.

It is suggested that during primary creep, material strain hardens thus increases its creep
resistance. Constant creep rate during secondary creep is believed to be due to balance
between the competing processes of strain hardening and recovery. The average value of
creep rate during the secondary creep is called the minimum creep rate. Third stage creep
occurs in constant load tests at high stresses at high temperatures. This stage is greatly
delayed in constant stress tests. Tertiary creep is believed to occur because of either
reduction in cross-sectional area due to necking or internal void formation. Third stage is
often associated with metallurgical changes such as coarsening of precipitate particles,
recrystallization, or diffusional changes in the phases that are present.

For metallic materials most creep tests are conducted in uni-axial tensile mode. However,
uni-axial compression tests are used for brittle materials to avoid stress amplification and
corresponding crack propagation. For most materials creep properties are independent of
loading direction.

The minimum creep rate is the most important design parameter derived from the creep
curve. It is the engineering design parameter that is considered for long-life applications,
e.g. nuclear power plant components. On the other hand, for short-life components (e.g.
turbine blades, rocket motor nozzles), time to rupture or rupture lifetime is the dominant
factor in design. It is found from test conducted to the point of failure, and the test is
known as stress-rupture test. This is basically similar to a creep test except that the test is
always carried out at higher loads. In a creep test total strain is often less than 0.5%,
while in stress-rupture test it is around 50%.

8.5.2 Stress and temperature effects

Creep properties of a material are usually characterized at a constant stress and a constant
temperature. A change in either of these parameters causes change in creep behavior of
the material. With increase in either stress or temperature (a) instantaneous elastic strain
increases (b) steady state creep rate increases and (c) rupture lifetime decreases.

Influence of stress on steady state creep rate can be written as

& s = K 1 n

where K1 and n are material constants. Value of these parameters can found from
logarithmic plot between creep rate and stress after conducting tests at a constant
temperature. Similarly temperature influence on steady state creep rate can be
characterized. Influence of combined action of temperature and stress on steady state
creep rate can be presented as follows:

Qc

& s = K 2 n e RT

where K2 and Qc (activation energy for creep) are constants. Values of n for a specific
material are dependent mechanism of creep operational in those particular conditions of
stress and temperature. Figure-8.16 presents the influence of stress/temperature on creep
behavior.

Figure-8.16: Influence of stress/temperature on creep behavior.

Creep data presented in form of stress-temperature diagrams are termed deformation

mechanism maps which indicates stress-temperature regimes over which various
mechanisms operate. The chief creep deformation mechanisms are:
 Dislocation glide involves dislocation moving along slip planes by overcoming
barrier with help of thermal activation. This occurs at high stresses, /G >10-2.

Dislocation creep involves movement of dislocations which overcome barriers

by thermally assisted mechanisms like diffusion of vacancies or interstitials.
Occurs at moderate stresses, 10-2 >/G >10-4.

Diffusion creep involves flow of vacancies and interstitials under the influence
of applied stress. Occurs for /G <10-4.

Grain boundary sliding involves sliding of grains against each other.

It is to be noted that, quite often, more than one creep mechanism will operate at a time.
When several mechanisms are in operation at a time simultaneously, i.e. independent of
each other, fastest mechanism will dominate the proceedings. On the other hand, when
they are operating in series, slowest mechanism will control the creep deformation.

References

1. G. E. Dieter, Mechanical Metallurgy, Third Edition, McGraw-Hill, New York,

1986.
2. R. W. Hertzberg, Deformation and Fracture Mechanics of Engineering Materials,
Fourth Edition, Wiley, New York, 1996.
3. T. H. Courtney, Mechanical Behavior of Materials, Second Edition, McGraw-
Hill, New York, 2000.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 9. Applications and Processing of Metals and Alloys

In the materials world we are living in, when making a new device/component, most
often we come across a very familiar problem. This is nothing but select the right
material. As learnt in earlier chapter, selection of material can play very important role
preventing failures. Selection of material for a specific purpose depends on many factors.
Some of the important ones are: strength, ease of forming, resistance to environmental
degradation, etc. Another dimension an engineer should be aware of it is how to tailor the
required properties of materials.

As introduced in one of the earlier chapters, materials can be are broadly classified as
metals, ceramics and plastics. This chapter introduces different classes of metallic
materials, common fabrication methods, and means to alter their properties on purpose.
Following chapters deal with ceramic materials and plastic materials.

9.1 Types of metals and alloys

Metallic materials are broadly of two kinds ferrous and non-ferrous materials. This
classification is primarily based on tonnage of materials used all around the world.
Ferrous materials are those in which iron (Fe) is the principle constituent. All other
materials are categorized as non-ferrous materials. Another classification is made based
on their formability. If materials are hard to form, components with these materials are
fabricated by casting, thus they are called cast alloys. If material can be deformed, they
are known as wrought alloys. Materials are usually strengthened by two methods cold
work and heat treatment. Strengthening by heat treatment involves either precipitation
hardening or martensitic transformation, both of which constitute specific heat treating
procedure. When a material can not be strengthened by heat treatment, it is referred as
non-heat-treatable alloys.

9.1.1 Ferrous materials

Ferrous materials are produced in larger quantities than any other metallic material. Three
factors account for it: (a) availability of abundant raw materials combined with
economical extraction, (b) ease of forming and (c) their versatile mechanical and physical
properties. One main drawback of ferrous alloys is their environmental degradation i.e.
poor corrosion resistance. Other disadvantages include: relatively high density and
comparatively low electrical and thermal conductivities. In ferrous materials the main
alloying element is carbon (C). Depending on the amount of carbon present, these alloys
will have different properties, especially when the carbon content is either less/higher
than 2.14%. This amount of carbon is specific as below this amount of carbon, material
undergoes eutectoid transformation, while above that limit ferrous materials undergo
eutectic transformation. Thus the ferrous alloys with less than 2.14% C are termed as
steels, and the ferrous alloys with higher than 2.14% C are termed as cast irons.

9.1.1.a Steels

Steels are alloys of iron and carbon plus other alloying elements. In steels, carbon present
in atomic form, and occupies interstitial sites of Fe microstructure. Alloying additions are
necessary for many reasons including: improving properties, improving corrosion
resistance, etc. Arguably steels are well known and most used materials than any other
materials.

Mechanical properties of steels are very sensitive to carbon content. Hence, it is practical
to classify steels based on their carbon content. Thus steels are basically three kinds: low-
carbon steels (% wt of C < 0.3), medium carbon steels (0.3 <% wt of C < 0.6) and high-
carbon steels (% wt of C > 0.6). The other parameter available for classification of steels
is amount of alloying additions, and based on this steels are two kinds: (plain) carbon
steels and alloy-steels.

Low carbon steels: These are arguably produced in the greatest quantities than other
alloys. Carbon present in these alloys is limited, and is not enough to strengthen these
materials by heat treatment; hence these alloys are strengthened by cold work. Their
microstructure consists of ferrite and pearlite, and these alloys are thus relatively soft,
ductile combined with high toughness. Hence these materials are easily machinable and
weldable. Typical applications of these alloys include: structural shapes, tin cans,
automobile body components, buildings, etc.

A special group of ferrous alloys with noticeable amount of alloying additions are known
as HSLA (high-strength low-alloy) steels. Common alloying elements are: Cu, V, Ni, W,
Cr, Mo, etc. These alloys can be strengthened by heat treatment, and yet the same time
they are ductile, formable. Typical applications of these HSLA steels include: support
columns, bridges, pressure vessels.

Medium carbon steels: These are stronger than low carbon steels. However these are of
less ductile than low carbon steels. These alloys can be heat treated to improve their
strength. Usual heat treatment cycle consists of austenitizing, quenching, and tempering
at suitable conditions to acquire required hardness. They are often used in tempered
condition. As hardenability of these alloys is low, only thin sections can be heat treated
using very high quench rates. Ni, Cr and Mo alloying additions improve their
hardenability. Typical applications include: railway tracks & wheels, gears, other
machine parts which may require good combination of strength and toughness.

High carbon steels: These are strongest and hardest of carbon steels, and of course their
ductility is very limited. These are heat treatable, and mostly used in hardened and
tempered conditions. They possess very high wear resistance, and capable of holding
sharp edges. Thus these are used for tool application such as knives, razors, hacksaw
blades, etc. With addition of alloying element like Cr, V, Mo, W which forms hard
carbides by reacting with carbon present, wear resistance of high carbon steels can be
improved considerably.

Stainless steels: The name comes from their high resistance to corrosion i.e. they are rust-
less (stain-less). Steels are made highly corrosion resistant by addition of special alloying
elements, especially a minimum of 12% Cr along with Ni and Mo. Stainless steels are
mainly three kinds: ferritic & hardenable Cr steels, austenitic and precipitation
hardenable (martensitic, semi-austenitic) steels. This classification is based on prominent
constituent of the microstructure. Typical applications include cutlery, razor blades,
surgical knives, etc.

Ferritic stainless steels are principally Fe-Cr-C alloys with 12-14% Cr. They also contain
small additions of Mo, V, Nb, and Ni.

Austenitic stainless steels usually contain 18% Cr and 8% Ni in addition to other minor
alloying elements. Ni stabilizes the austenitic phase assisted by C and N. Other alloying
additions include Ti, Nb, Mo (prevent weld decay), Mn and Cu (helps in stabilizing
austenite).

By alloying additions, for martensitic steels Ms is made to be above the room

temperature. These alloys are heat treatable. Major alloying elements are: Cr, Mn and
Mo.

Ferritic and austenitic steels are hardened and strengthened by cold work because they
are not heat treatable. On the other hand martensitic steels are heat treatable. Austenitic
steels are most corrosion resistant, and they are produced in large quantities. Austenitic
steels are non-magnetic as against ferritic and martensitic steels, which are magnetic.

9.1.1.b Cast irons

Though ferrous alloys with more than 2.14 wt.% C are designated as cast irons,
commercially cast irons contain about 3.0-4.5% C along with some alloying additions.
Alloys with this carbon content melt at lower temperatures than steels i.e. they are
responsive to casting. Hence casting is the most used fabrication technique for these
alloys.

Hard and brittle constituent presented in these alloys, cementite is a meta-stable phase,
and can readily decompose to form -ferrite and graphite. In this way disadvantages of
brittle phase can easily be overcome. Tendency of cast irons to form graphite is usually
controlled by their composition and cooling rate. Based on the form of carbon present,
cast irons are categorized as gray, white, nodular and malleable cast irons.

Gray cast iron: These alloys consists carbon in form graphite flakes, which are
surrounded by either ferrite or pearlite. Because of presence of graphite, fractured surface
of these alloys look grayish, and so is the name for them. Alloying addition of Si (1-
3wt.%) is responsible for decomposition of cementite, and also high fluidity. Thus
castings of intricate shapes can be easily made. Due to graphite flakes, gray cast irons are
weak and brittle. However they possess good damping properties, and thus typical
applications include: base structures, bed for heavy machines, etc. they also show high
resistance to wear.

White cast iron: When Si content is low (< 1%) in combination with faster cooling rates,
there is no time left for cementite to get decomposed, thus most of the brittle cementite
retains. Because of presence of cementite, fractured surface appear white, hence the
name. They are very brittle and extremely difficult to machine. Hence their use is limited
to wear resistant applications such as rollers in rolling mills. Usually white cast iron is
heat treated to produce malleable iron.

Nodular (or ductile) cast iron: Alloying additions are of prime importance in producing
these materials. Small additions of Mg / Ce to the gray cast iron melt before casting can
result in graphite to form nodules or sphere-like particles. Matrix surrounding these
particles can be either ferrite or pearlite depending on the heat treatment. These are
stronger and ductile than gray cast irons. Typical applications include: pump bodies,
crank shafts, automotive components, etc.

Malleable cast iron: These formed after heat treating white cast iron. Heat treatments
involve heating the material up to 800-900 C, and keep it for long hours, before cooling
it to room temperature. High temperature incubation causes cementite to decompose and
form ferrite and graphite. Thus these materials are stronger with appreciable amount of
ductility. Typical applications include: railroad, connecting rods, marine and other heavy-
duty services.

9.1.2 Non-ferrous materials

Non-ferrous materials have specific advantages over ferrous materials. They can be
fabricated with ease, high relatively low density, and high electrical and thermal
conductivities. However different materials have distinct characteristics, and are used for
specific purposes. This section introduces some typical non-ferrous metals and their
alloys of commercial importance.

Aluminium alloys: These are characterized by low density, high thermal & electrical
conductivities, and good corrosion resistant characteristics. As Al has FCC crystal
structure, these alloys are ductile even at low temperatures and can be formed easily.
However, the great limitation of these alloys is their low melting point (660 C), which
restricts their use at elevated temperatures. Strength of these alloys can be increased by
both cold and heat treatment based on these alloys are designated in to two groups, cast
and wrought. Chief alloying elements include: Cu, Si, Mn, Mg, Zn. Recently, alloys of Al
and other low-density metals like Li, Mg, Ti gained much attention as there is much
concern about vehicle weight reduction. Al-Li alloys enjoy much more attention
especially as they are very useful in aircraft and aerospace industries. Common
applications of Al alloys include: beverage cans, automotive parts, bus bodies, aircraft
structures, etc. Some of the Al alloys are capable of strengthening by precipitation, while
others have to be strengthened by cold work or solid solution methods.

Copper alloys: As history goes by, bronze has been used for thousands of years. It is
actually an alloy of Cu and Sn. Unalloyed Cu is soft, ductile thus hard to machine, and
has virtually unlimited capacity for cold work. One special feature of most of these alloys
is their corrosion resistant in diverse atmospheres. Most of these alloys are strengthened
by either cold work or solid solution method. Common most Cu alloys: Brass, alloys of
Cu and Zn where Zn is substitutional addition (e.g.: yellow brass, catridge brass, muntz
metal, gilding metal); Bronze, alloys of Cu and other alloying additions like Sn, Al, Si
and Ni. Bronzes are stronger and more corrosion resistant than brasses. Mention has to be
made about Beryllium coppers who possess combination of relatively high strength,
excellent electrical and corrosion properties, wear resistance, can be cast, hot worked and
cold worked. Applications of Cu alloys include: costume jewelry, coins, musical
instruments, electronics, springs, bushes, surgical and dental instruments, radiators, etc.

Magnesium alloys: The most sticking property of Mg is its low density among all
structural metals. Mg has HCP structure, thus Mg alloys are difficult to form at room
temperatures. Hence Mg alloys are usually fabricated by casting or hot working. As in
case of Al, alloys are cast or wrought type, and some of them are heat treatable. Major
alloying additions are: Al, Zn, Mn and rare earths. Common applications of Mg alloys
include: hand-held devices like saws, tools, automotive parts like steering wheels, seat
frames, electronics like casing for laptops, camcoders, cell phones etc.

Titanium alloys: Ti and its alloys are of relatively low density, high strength and have
very high melting point. At the same time they are easy to machine and forge. However
the major limitation is Tis chemical reactivity at high temperatures, which necessitated
special techniques to extract. Thus these alloys are expensive. They also possess excellent
corrosion resistance in diverse atmospheres, and wear properties. Common applications
include: space vehicles, airplane structures, surgical implants, and petroleum & chemical
industries.

Refractory metals: These are metals of very high melting points. For example: Nb, Mo,
W and Ta. They also possess high strength and high elastic modulus. Common
applications include: space vehicles, x-ray tubes, welding electrodes, and where there is a
need for corrosion resistance.

Noble metals: These are eight all together: Ag, Au, Pt, Pa, Rh, Ru, Ir and Os. All these
possess some common properties such as: expensive, soft and ductile, oxidation resistant.
Ag, Au and Pt are used extensively in jewelry, alloys are Ag and Au are employed as
dental restoration materials; Pt is used in chemical reactions as a catalyst and in thermo
couples.

9.2 Fabrication of metals

Metals are fabricated by different means to achieve metals and alloys of desired
characteristics. There been many kinds of fabrication techniques, and for a particular
metal use of these depend on properties of metal, product shape-size-properties, cost, etc.
Metal fabrication techniques are mainly four kinds: Casting - to give a shape by pouring
in liquid metal into a mold that holds the required shape, and letting harden the metal
without external pressure; Forming to give shape in solid state by applying pressure;
Machining in which material is removed in order to give it the required shape; and
Joining where different parts are joined by various means. One of the most important
miscellaneous techniques is powder metallurgy.

9.2.1 Metal casting

This technique is employed when (a) product is large and/or complicated shape (b)
particular material is low in ductility. This is also employed as it is usually economical
compared with other techniques. Different casting techniques include: sand, die,
investment, continuous casting

Sand casting: The common casting method where sand is used as casting material. A two
piece mold (cope and drag) is formed by compact packing of sand around a pattern of
required shape. An addition gating is provided for proper distribution of liquid metal.

Die casting: Here metal is forced into mold by external pressure at high velocities.
Usually a permanent two-piece mold made of steel is used. In this technique rapid
cooling rates are achieved, thus inexpensive.

Investment casting: In this pattern is made of wax. Then fluid slurry of casting material is
poured over which eventually hardens and holds the required shape. Subsequently,
pattern material is heated to leave behind the cavity. This technique is employed when
high dimensional accuracy, reproduction of fine details, and an excellent finish are
required. For example: jewelry, dental crowns, and gas turbine blades jet engine
impellers.

Continuous casting: After refining metals are usually in molten state, which are later
solidified into ingots for further processing like forming. In continuous casting,
solidification and primary forming process are combined, where refined metal is cast
directly into a continuous strand which is cooled by water jets. This technique is highly
automated and more efficient. Uniform composition through-out the casting is achievable
when compared with ingot-cast products.

9.2.2 Metal forming

In these techniques, a metallic piece is subjected to external pressures (in excess of yield
strength of the material) to induce deformation, thus material acquires a desired shape.
These are basically two types one that performed at relatively low temperatures, cold
working; ad the other performed at high temperatures, hot working. Hot working is
responsible mainly for substantial change in cross section without material getting
strengthened, while during cold working, fine details are achieved along with material
getting strengthened. Most common forming techniques are: forging, rolling, extrusion,
and drawing. Figure-9.1 illustrates different forming processes.

Figure-9.1: Different forming processes.

Forging: This involves deforming a single piece of metal, usually, by successive blows or
continuous squeezing. In open die forging, two dies having same shape is employed,
usually, over large work-pieces; while in closed die forging, there may be more than two
pieces of die put together having finished shape. Forged products have outstanding grain
structures and very good mechanical properties. Typical products include: crane hooks,
wrenches, crank shafts, connecting rods.

Rolling: Most widely used forming technique because of high production rate and close
dimensional control of final product. It involves passing a piece of metal between two
rotating rolls. Deformation is terms of reduction in thickness resulting from applied
compressive forces. This technique is typically employed to produce sheets, strips, foil, I-
beams, rails, etc.

Extrusion: In this technique a piece of material is forced though a die orifice by a

compressive force. Final product emerging from die will have the desired shape and
reduced cross sectional area, and will constant cross-section over very long lengths. Two
varieties of extrusion are direct extrusion and indirect extrusion, where distinction limits
to movement of tool and final product and consequent changes in required force. Typical
extrusion products are: rods, (seamless) tubes, complicated shapes for domestic purpose.

Drawing: It is pulling of material though die orifice using tensile forces. Again a
reduction in cross-section results with corresponding change in length. Drawing die
entrance is at angle against to extrusion die which is usually rectangular. Typical drawing
strand includes number of dies in a series sequence. Rods, wire, and tubes are commonly
produced using drawing technique.

9.2.3 Machining

This technique employs removable of metal from selected areas of the workpiece to give
final shape to the product. This is in direct contrast with metal forming where metal is
moved and volume is conserved. Machining usually is employed to produce shapes with
high dimensional tolerance, good surface finish, and often with complex geometry. And
another important note is that when number of product pieces required is small,
machining is preferred over forming as special tool cost will be less.

9.2.4 Joining

There been many joining techniques, especially for metallic materials. These include:
welding, brazing, soldering, and riveting. In these techniques, two pieces are joined
together either by adhesive/cohesive bonding and/or mechanical locking. Welding,
brazing, and soldering involve melting of either parent metal or external metallic liquid
(filler material) which upon cooling provides cohesive bonds. In riveting, pieces are put
together by mechanical locking. These techniques are employed to join two pieces of
same metal with complicated shapes, or of different metals because of difficulty in
fabricating them using one of the previous methods. This may be employed when on-part
fabrication is expensive or inconvenient.

9.2.5 Powder metallurgy

In this technique, metal powders or mixture of metal powers at desired relative amounts
are compacted into the desired shape, followed by sintering in controlled atmosphere to
produce a denser product. It is makes it possible to produce a virtually non-porous
product where diffusional processes controls the efficiency of the process. It is suitable,
especially, for metals with low ductility/high melting points. Other advantage include:
close dimensional tolerance of complicated shapes. Usually products are less dense than
wrought products because of porosity. However, it is advantageous as pores can retain oil
for self-lubrication of bushes, and high damping capacity.

9.3 Thermal processing of metals and alloys

Apart from mechanical processing, metals are very often subjected to thermal processing
for various reasons, like: to refine grain structure/size, to minimize residual stresses, to
impart phase changes, to develop special phases over external surfaces, etc. Metals and
alloys develop requisite properties by thermal processing either through grain refinement
of phase changes. Thermal processing is also known as heat treatment. Heat treatment
originated as an ancient art in mans attempts to improve the performance of materials in
their practical applications. In present day metallurgical practice, heat treatment has
become very important for obvious reasons. There has been tremendous progress over
centuries in the systematic understanding of materials structure and structure-property
relationships that eliminated the empiricism in thermal processing. Properly designed and
implemented thermal processing can result in optimum modifications in the composition
and distribution of phases, corresponding changes in physical, chemical and mechanical
properties at substantial levels. However, most of the thermal processes are aimed to
improving mechanical characteristics of materials. Thus it is possible to extend the
service performance of materials considerably within constraints of available resources.

All metals can be subjected to thermal processing. But the effect of it may differ from
one metal to another. Metals are subjected to heat treatment for one or more of the
following purposes: improvement in ductility; relieving internal stresses; grain size
refinement; increase of strength; improvement in machinability, toughness; etc.

Heat treatment of materials involves number of factors temperature up to which

material is heated, length of time that the material is held at the elevated temperature, rate
of cooling, and the surrounding atmosphere under the thermal treatment. All these factors
depend on material, pre-processing of the materials chemical composition, size and
shape of the object, final properties desired, materials melting point/liquidus, etc.

Thermal processes may be broadly classified into two categories based on cooling rates
from elevated temperatures annealing and quenching & tempering. Annealing involved
cooling the material from elevated temperatures slowly, while quenching means very fast
cooling of the material using cooling medium like water/oil bath. Quenching is done to
retain the phases of elevated temperatures at room temperature.

9.3.1 Annealing processes

The term annealing was used by craftsmen who discovered the beneficial effects of
heating the material at elevated temperatures followed by slow cooling of it to room
temperature. Annealing can be defined as a heat treatment process in which the material
is taken to a high temperature, kept there for some time and then cooled. High
temperatures allow diffusion processes to occur fast. The time at the high temperature
(soaking time) must be long enough to allow the desired transformation to occur. Cooling
is done slowly to avoid the distortion (warping) of the metal piece, or even cracking,
caused by stresses induced by differential contraction due to thermal inhomogeneities.
Benefits of annealing are:

relieve stresses
increase softness, ductility and toughness
produce a specific microstructure
Depending on the specific purpose, annealing is classified into various types: process
annealing, stress relief, full annealing and normalizing.

Process annealing is primarily applied to cold worked metals to negate the effects of cold
work. During this heat treatment, material becomes soft and thus its ductility will be
increased considerably. It is commonly sandwiched between two cold work operations.
During this, recovery and recrystallization are allowed whereas grain growth was
restricted.

Stress relief operation removes the stresses that might have been generated during plastic
deformation, non-uniform cooling, or phase transformation. Unless removed, these
stresses may cause distortion of components. Temperature used is normally low such that
effects resulting from cold working are not affected.

Full annealing is normally used for products that are to be machined subsequently, such
as transmission gear blanks. After heating and keeping at an elevated temperature,
components are cooled in furnace to effect very slow cooling rates. Typically, the product
receives additional heat treatments after machining to restore hardness and strength.

Normalizing is used to refine the grains and produce a more uniform and desirable size
distribution. It involves heating the component to attain single phase (e.g.: austenite in
steels), then cooling in open air atmosphere.

9.3.2 Quenching and Tempering processes

Quenching is heat treatment process where material is cooled at a rapid rate from
elevated temperature to produce Martensite phase. This process is also known as
hardening. Rapid cooling rates are accomplished by immersing the components in a
quench bath that usually contains quench media in form of either water or oil,
accompanied by stirring mechanism.

Quenching process is almost always followed by tempering heat treatment. Tempering is

the process of heating martensitic steel at a temperature below the eutectoid
transformation temperature to make it softer and more ductile. During the tempering
process, Martensite transforms to a structure containing iron carbide particles in a matrix
of ferrite.

Martempering is a modified quenching procedure used to minimize distortion and

cracking that may develop during uneven cooling of the heat-treated material. It involves
cooling the austenized steel to temperature just above Ms temperature, holding it there
until temperature is uniform, followed by cooling at a moderate rate to room temperature
before austenite-to-bainite transformation begins. The final structure of martempered
steel is tempered Martensite.

Austempering is different from martempering in the sense that it involves austenite-to-

bainite transformation. Thus, the structure of austempered steel is bainite. Advantages of
austempering are improved ductility; decreased distortion and disadvantages are need
for special molten bath; process can be applied to limited number of steels.

9.4 Case Hardening

In case hardening, the surface of the steel is made hard and wear resistant, but the core
remains soft and tough. Such a combination of properties is desired in applications such
as gears.

9.4.1. Induction hardening

Here, an alternating current of high frequency passes through an induction coil enclosing
the steel part to be heat treated. The induced emf heats the steel. The depth up to which
the heat penetrates and raises the temperature above the elevated temperature is inversely
proportional to the square root of the ac frequency. In induction hardening, the heating
time is usually a few seconds. Immediately after heating, water jets are activated to
quench the surface. Martensite is produced at the surface, making it hard and wear
resistant. The microstructure of the core remains unaltered. Induction hardening is
suitable for mass production of articles of uniform cross-section.

9.4.2. Flame hardening

For large work pieces and complicated cross-sections induction heating is not easy to
apply. In such cases, flame hardening is done by means of an oxyacetylene torch.
Heating should be done rapidly by the torch and the surface quenched, before appreciable
heat transfer to the core occurs

9.4.3. Laser hardening

In this case, a laser beam can be used for surface hardening. As laser beams are of high
intensity, a lens is used to reduce the intensity by producing a defocused spot of size
ranging from 0.5 to 25 mm. Proper control of energy input is necessary to avoid melting.
Laser hardening has the advantage of precise control over the area to be hardened, an
ability to harden reentrant surfaces, very high speed of hardening and no separate
quenching step. The disadvantage is that the hardening is shallower than in induction and
flame hardening

9.4.4. Carburizing

Carburizing is the most widely used method of surface hardening. Here, the surface
layers of low carbon steel are enriched with carbon up to 0.8-1.0%. The source of carbon
may be a solid medium, a liquid or a gas. In all cases, the carbon enters the steel at the
surface and diffuses into the steel as a function of time at an elevated temperature.
Carburizing is done at 920-950o C. This fully austenitic state is essential. If carburizing
is done in the ferritic region, the carbon, with very limited solubility in ferrite, tends to
form massive cementite particles near the surface, making the subsequent heat treatment
difficult. For this reason, carburizing is always done in the austenitic state, even though
longer times are required due to the diffusion rate of carbon in austenite being less that in
ferrite at such temperatures.

9.4.5. Cyaniding

Cyaniding is done in a liquid bath of NaCN, with the concentration varying between 30
and 97%. The temperature used for cyaniding is lower than that for carburizing and is in
the range of 800-870o C. The time of cyaniding is 1-3 hr to produce a case depth of 0.25
mm or less

9.4.5. Nitriding

Nitriding is carried out in the ferritic region. No phase change occurs after nitriding. The
part to be nitrided should posses the required core properties prior to nitriding. During
nitriding, pure ammonia decomposes to yield nitrogen which enters the steel. The
solubility of nitrogen in ferrite is small. Most of the nitrogen, that enters the steel, forms
hard nitrides (e.g., Fe3N). The temperature of nitriding is 500-590o C. The time for a
case depth of 0.02 mm is about 2 hr. In addition to providing outstanding wear resistance,
the nitride layer increases the resistance of carbon steel to corrosion in moist
atmospheres.

References

1. G. E. Dieter, Mechanical Metallurgy, Third Edition, McGraw-Hill, New York,

1986.
2. William D. Callister, Jr, Materials Science and Engineering An introduction,
sixth edition, John Wiley & Sons, Inc. 2004.
3. ASM Handbook, Heat treating, Vol. 4, ASM International, Materials Park, OH,
1991.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 10. Applications and Processing of Ceramics

Ceramics form an important part of materials group. Ceramics are compounds between
metallic and nonmetallic elements for which the inter-atomic bonds are either ionic or
predominantly ionic. The term ceramics comes from the Greek word keramikos which
means burnt stuff. Characteristic properties of ceramics are, in fact, optimized through
thermal treatments. They exhibit physical properties those are different from that of
metallic materials. Thus metallic materials, ceramics, and even polymers tend to
complement each other in service.

10.1 Types and applications of ceramics

Ceramics greatly differ in their basic composition. The properties of ceramic materials
also vary greatly due to differences in bonding, and thus found a wide range of
engineering applications. Classification of ceramics based on their specific applications
and composition are two most important ways among many.

Based on their composition, ceramics are classified as:

Oxides,
Carbides,
Nitrides,
Sulfides,
Fluorides, etc.

The other important classification of ceramics is based on their application, such as:

Glasses,
Clay products,
Refractories,
Abrasives,
Cements,
 Advanced ceramics.

In general, ceramic materials used for engineering applications can be divided into two
groups: traditional ceramics, and the engineering ceramics. Typically, traditional
ceramics are made from three basic components: clay, silica (flint) and feldspar. For
example bricks, tiles and porcelain articles. However, engineering ceramics consist of
highly pure compounds of aluminium oxide (Al2O3), silicon carbide (SiC) and silicon
nitride (Si3N4).

Glasses: glasses are a familiar group of ceramics containers, windows, mirrors, lenses,
etc. They are non-crystalline silicates containing other oxides, usually CaO, Na2O, K2O
and Al2O3 which influence the glass properties and its color. Typical property of glasses
that is important in engineering applications is its response to heating. There is no
definite temperature at which the liquid transforms to a solid as with crystalline materials.
A specific temperature, known as glass transition temperature or fictive temperature is
defined based on viscosity above which material is named as super cooled liquid or
liquid, and below it is termed as glass.

Clay products: clay is the one of most widely used ceramic raw material. It is found in
great abundance and popular because of ease with which products are made. Clay
products are mainly two kinds structural products (bricks, tiles, sewer pipes) and white-
wares (porcelain, chinaware, pottery, etc.).

Refractories: these are described by their capacity to withstand high temperatures without
melting or decomposing; and their inertness in severe environments. Thermal insulation
is also an important functionality of refractories.

Abrasive ceramics: these are used to grind, wear, or cut away other material. Thus the
prime requisite for this group of materials is hardness or wear resistance in addition to
high toughness. As they may also exposed to high temperatures, they need to exhibit
some refractoriness. Diamond, silicon carbide, tungsten carbide, silica sand, aluminium
oxide / corundum are some typical examples of abrasive ceramic materials.

Cements: cement, plaster of paris and lime come under this group of ceramics. The
characteristic property of these materials is that when they are mixed with water, they
form slurry which sets subsequently and hardens finally. Thus it is possible to form
virtually any shape. They are also used as bonding phase, for example between
construction bricks.

Advanced ceramics: these are newly developed and manufactured in limited range for
specific applications. Usually their electrical, magnetic and optical properties and
combination of properties are exploited. Typical applications: heat engines, ceramic
armors, electronic packaging, etc.

Some typical ceramics and respective applications are as follows:

Aluminium oxide / Alumina (Al2O3): it is one of most commonly used ceramic material. It
is used in many applications such as to contain molten metal, where material is operated
at very high temperatures under heavy loads, as insulators in spark plugs, and in some
unique applications such as dental and medical use. Chromium doped alumina is used for
making lasers.

Aluminium nitride (AlN): because of its typical properties such as good electrical
insulation but high thermal conductivity, it is used in many electronic applications such
as in electrical circuits operating at a high frequency. It is also suitable for integrated
circuits. Other electronic ceramics include barium titanate (BaTiO3) and Cordierite
(2MgO-2Al2O3-5SiO2).

Diamond (C): it is the hardest material known to available in nature. It has many
applications such as industrial abrasives, cutting tools, abrasion resistant coatings, etc. it
is, of course, also used in jewelry.

Lead zirconium titanate (PZT): it is the most widely used piezoelectric material, and is
used as gas igniters, ultrasound imaging, in underwater detectors.

Silica (SiO2): is an essential ingredient in many engineering ceramics, thus is the most
widely used ceramic material. Silica-based materials are used in thermal insulation,
abrasives, laboratory glassware, etc. it also found application in communications media
as integral part of optical fibers. Fine particles of silica are used in tires, paints, etc.

Silicon carbide (SiC): it is known as one of best ceramic material for very high
temperature applications. It is used as coatings on other material for protection from
extreme temperatures. It is also used as abrasive material. It is used as reinforcement in
many metallic and ceramic based composites. It is a semiconductor and often used in
high temperature electronics. Silicon nitride (Si3N4) has properties similar to those of SiC
but is somewhat lower, and found applications in such as automotive and gas turbine
engines.

Titanium oxide (TiO2): it is mostly found as pigment in paints. It also forms part of
certain glass ceramics. It is used to making other ceramics like BaTiO3.

Titanium boride (TiB2): it exhibits great toughness properties and hence found
applications in armor production. It is also a good conductor of both electricity and heat.

Uranium oxide (UO2): it is mainly used as nuclear reactor fuel. It has exceptional
dimensional stability because its crystal structure can accommodate the products of
fission process.

Yttrium aluminium garnet (YAG, Y3Al5O12): it has main application in lasers (Nd-YAG
lasers).
Zirconia (ZrO2): it is also used in producing many other ceramic materials. It is also used
in making oxygen gas sensors, as additive in many electronic ceramics. Its single crystals
are part of jewelry.

10.2 Fabrication and processing of ceramics

Ceramics melt at high temperatures and they exhibit a brittle behavior under tension. As a
result, the conventional melting, casting and thermo-mechanical processing routes are not
suitable to process the polycrystalline ceramics. Inorganic glasses, though, make use of
lower melting temperatures due to formation of eutectics. Hence, most ceramic products
are made from ceramic powders through powder processing starting with ceramic
powders. The powder processing of ceramics is very close to that of metals, powder
metallurgy. However there is an important consideration in ceramic-forming that is more
prominent than in metal forming: it is dimensional tolerance. Post forming shrinkage is
much higher in ceramics processing because of the large differential between the final
density and the as-formed density.

Glasses, however, are produced by heating the raw materials to an elevated temperature
above which melting occurs. Most commercial glasses are of the silica-soda-lime variety,
where silica is supplied in form of common quartz sand, soda (Na2O) in form of soda ash
(Na2CO3) while the lime (CaO) is supplied in form of limestone (CaCO3). Different
forming methods- pressing, blowing, drawing and fiber forming- are widely in practice to
fabricate glass products. Thick glass objects such as plates and dishes are produced by
pressing, while the blowing is used to produce objects like jars, bottles and light bulbs.
Drawing is used to form long objects like tubes, rods, fibers, whiskers etc. The pressing
and blowing process is shown in figure 10.1

Figure 10.1: Schematic diagram of pressing and blowing processes

Ceramic powder processing consists of powder production by milling/grinding, followed
by fabrication of green product, which is then consolidated to obtain the final product. A
powder is a collection of fine particles. Synthesis of powder involves getting it ready for
shaping by crushing, grinding, separating impurities, blending different powders, drying
to form soft agglomerates. Different techniques such as compaction, tape casting, slip
casting, injection molding and extrusion are then used to convert processed powders into
a desired shape to form what is known as green ceramic. The green ceramic is then
consolidated further using a high-temperature treatment known as sintering or firing.

As-mined raw materials are put through a milling or grinding operation in which particle
size is reduced to and physically liberate the minerals of interest from the rest of the
gangue material. Wet milling is much more common with ceramic materials than with
metals. The combination of dry powders with a dispersant such as water is called slurry.
Ball- and vibratory- milling is employed to further reduce the size of minerals and to
blend different powders.

Ceramic powders prepared are shaped using number of techniques, such as casting,
compaction, extrusion/hydro-plastic forming, injection molding. Tape casting, also
known as doctor blade process, is used for the production of thin ceramic tapes. The
schematic diagram of tape casting process is shown in figure 10.2. In this technique
slurry containing ceramic particles, solvent, plasticizers, and binders is then made to flow
under a blade and onto a plastic substrate. The shear thinning slurry spreads under the
blade. The tape is then dried using clean hot air. Later-on the tape is subjected to binder
burnout and sintering operations. Tape thickness normally range between 0.1 and 2 mm.
Commercially important electronic packages based on alumina substrates and barium
titanate capacitors are made using this technique. A schematic diagram of doctor blade
process is shown in the figure.

Figure 10.2: Schematic diagram of tape casting process

Slip casting is another casting technique widely used. This technique uses aqueous
slurry, also known as slip, of ceramic powder. The slip is poured into a plaster of Paris
(CaSO4:2H2O) mold. As the water from slurry begins to move out by capillary action, a
thick mass builds along the mold wall. When sufficient product thickness is built, the rest
of the slurry is poured out (drain casting). It is also possible to continue to pour more
slurry in to form a solid piece (solid casting). The schematic diagram of slip casting
process is shown in figure 10.3

Figure 10.3: Schematic diagram of slip casting process

Extrusion and injection molding techniques are used to make products like tubes, bricks,
tiles etc. The basis for extrusion process is a viscous mixture of ceramic particles, binder
and other additives, which is fed through an extruder where a continuous shape of green
ceramic is produced. The product is cut to required lengths and then dried and sintered.
Injection molding of ceramics is similar to that of polymers. Ceramic powder is mixed
with a plasticizer, a thermoplastic polymer, and additives. Then the mixture is injected
into a die with use of an extruder. The polymer is then burnt off and the rest of the
ceramic shape is sintered at suitable high temperatures. Ceramic injection molding is
suitable for producing complex shapes. Figure 10.4 shows schematically the injection
molding process

Figure 10.4: Schematic diagram of Injection molding

Most popular technique to produce relatively simple shapes of ceramic products in large
numbers is combination of compaction and sintering. For example: electronic ceramics,
magnetic ceramics, cutting tools, etc. Compaction process is used to make green
ceramics that have respectable strength and can be handled and machined. Time for
compaction process varies from within a minute to hours depending on the complexity
and size of the product. Basically compaction process involves applying equal pressure in
all directions to a mixture ceramic powder to increase its density. In some cases,
compaction involves application of pressure using oil/fluid at room temperatures, called
cold iso-static pressing (CIP). Then the green ceramic is sintered with or without
pressure. CIP is used to achieve higher ceramic density or where the compaction of more
complex shapes is required. In some instances, parts may be produced under conditions
in which compaction and sintering are conducted under pressure at elevated temperatures.
This technique is known as hot iso-static pressing (HIP), and is used for refractory and
covalently bonded ceramics that do not show good bonding characteristics under CIP.
HIP is also used when close to none porosity is the requirement. Another characteristic
feature of HIP is high densities can be achieved without appreciable grain growth.

Sintering is the firing process applied to green ceramics to increase its strength. Sintering
is carried out below the melting temperature thus no liquid phase presents during
sintering. However, for sintering to take place, the temperature must generally be
maintained above one-half the absolute melting point of the material. During sintering,
the green ceramic product shrinks and experiences a reduction in porosity. This leads to
an improvement in its mechanical integrity. These changes involve different mass
transport mechanisms that cause coalescence of powder particles into a more dense mass.
With sintering, the grain boundary and bulk atomic diffusion contribute to densification,
surface diffusion and evaporation condensation can cause grain growth, but do not cause
densification. After pressing, ceramic particles touch one another. During initial stages of
sintering, necks form along the contact regions between adjacent particles thus every
interstice between particles becomes a pore. The pore channels in the compact grow in
size, resulting in a significant increase in strength. With increase in sintering time, pores
become smaller in size. The driving force for sintering process is the reduction in total
particle surface area, and thus the reduction in total surface energy. During sintering,
composition, impurity control and oxidation protection are provided by the use of
vacuum conditions or inert gas atmospheres.

References

1. W. D. Kingery, H. K. Bowen, and D. R. Uhlmann, Introduction to Ceramics,

Second Edition, Wiley, New York, 1976.
2. L. H. Van Vlack, Physical Ceramics for Engineers, Addison-Wesley Longman,
Reading, MA, 1964.
3. William D. Callister, Jr, Materials Science and Engineering An introduction,
sixth edition, John Wiley & Sons, Inc. 2004.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 11. Applications and Processing of Polymers

Polymers play a very important role in human life. In fact, our body is made of lot of
polymers, e.g. Proteins, enzymes, etc. Other naturally occurring polymers like wood,
rubber, leather and silk are serving the humankind for many centuries now. Modern
scientific tools revolutionized the processing of polymers thus available synthetic
polymers like useful plastics, rubbers and fiber materials. As with other engineering
materials (metals and ceramics), the properties of polymers are related their constituent
structural elements and their arrangement. The suffix in polymer mer is originated from
Greek word meros which means part. The word polymer is thus coined to mean
material consisting of many parts/mers. Most of the polymers are basically organic
compounds, however they can be inorganic (e.g. silicones based on Si-O network). This
chapter introduces classification of polymers, processing and synthesis of polymers,
followed by mechanism of deformation and mechanical behavior of polymers.

11.1 Polymer types and Polymer synthesis & processing

Polymers are classified in several ways by how the molecules are synthesized, by their
molecular structure, or by their chemical family. For example, linear polymers consist of
long molecular chains, while the branched polymers consist of primary long chains and
secondary chains that stem from these main chains. However, linear does not mean
straight lines. The better way to classify polymers is according to their mechanical and
thermal behavior. Industrially polymers are classified into two main classes plastics and
elastomers.

Plastics are moldable organic resins. These are either natural or synthetic, and are
processed by forming or molding into shapes. Plastics are important engineering
materials for many reasons. They have a wide range of properties, some of which are
unattainable from any other materials, and in most cases they are relatively low in cost.
Following is the brief list of properties of plastics: light weight, wide range of colors, low
thermal and electrical conductivity, less brittle, good toughness, good resistance to acids,
bases and moisture, high dielectric strength (use in electrical insulation), etc. Plastics are
again classified in two groups depending on their mechanical and thermal behavior as
thermoplasts (thermoplastic polymers) and thermosets (thermosetting polymers).

Thermoplasts: These plastics soften when heated and harden when cooled processes
that are totally reversible and may be repeated. These materials are normally fabricated
by the simultaneous application of heat and pressure. They are linear polymers without
any cross-linking in structure where long molecular chains are bonded to each other by
secondary bonds and/or inter-wined. They have the property of increasing plasticity with
increasing temperature which breaks the secondary bonds between individual chains.
Common thermoplasts are: acrylics, PVC, nylons, polypropylene, polystyrene,
polymethyl methacrylate (plastic lenses or perspex), etc.

Thermosets: These plastics require heat and pressure to mold them into shape. They are
formed into a permanent shape and cured or set by chemical reactions such as extensive
cross-linking. They cannot be re-melted or reformed into another shape but decompose
upon being heated to too high a temperature. Thus thermosets cannot be recycled,
whereas thermoplasts can be recycled. The term thermoset implies that heat is required to
permanently set the plastic. Most thermosets composed of long chains that are strongly
cross-linked (and/or covalently bonded) to one another to form 3-D network structures to
form a rigid solid. Thermosets are generally stronger, but more brittle than thermoplasts.
Advantages of thermosets for engineering design applications include one or more of the
following: high thermal stability, high dimensional stability, high rigidity, light weight,
high electrical and thermal insulating properties and resistance to creep and deformation
under load. There are two methods whereby cross-linking reaction can be initiated
cross-linking can be accomplished by heating the resin in a suitable mold (e.g. bakelite),
or resins such as epoxies (araldite) are cured at low temperature by the addition of a
suitable cross-linking agent, an amine. Epoxies, vulcanized rubbers, phenolics,
unsaturated polyester resins, and amino resins (ureas and melamines) are examples of
thermosets.

Elastomers: Also known as rubbers, these are polymers which can undergo large
elongations under load, at room temperature, and return to their original shape when the
load is released. There are number of man-made elastomers in addition to natural rubber.
These consist of coil-like polymer chains those can reversibly stretch by applying a force.

Processing of polymers mainly involves preparing a particular polymer by synthesis of

available raw materials, followed by forming into various shapes. Raw materials for
polymerization are usually derived from coal and petroleum products. The large
molecules of many commercially useful polymers must be synthesized from substances
having smaller molecules. The synthesis of the large molecule polymers is known as
polymerization in which monomer units are joined over and over to become a large
molecule. More upon, properties of a polymer can be enhanced or modified with the
addition of special materials. This is followed by forming operation. Addition
polymerization and condensation polymerization are the two main ways of
polymerization.
Addition polymerization, also known as chain reaction polymerization, is a process in
which multi-functional monomer units are attached one at a time in chainlike fashion to
form linear/3-D macro-molecules. The composition of the macro-molecule is an exact
multiple of for that of the original reactant monomer. This kind of polymerization
involves three distinct stages initiation, propagation and termination. To initiate the
process, an initiator is added to the monomer. This forms free radicals with a reactive site
that attracts one of the carbon atoms of the monomer. When this occurs, the reactive site
is transferred to the other carbon atom in the monomer and a chain begins to form in
propagation stage. A common initiator is benzoyl peroxide. When polymerization is
nearly complete, remaining monomers must diffuse a long distance to reach reactive site,
thus the growth rate decreases.

The process for polyethylene is as follows

Here R. represents the active initiator. Propagation involves the linear growth of the
molecule as monomer units become attached to one another in succession to produce the
chain molecule, which is represented, again for polyethylene, as follows

As we need polymers with controlled molecular weight, polymerization needs to be

terminated at some stage. Propagation may end or terminate in different ways. First, the
active ends of two propagating chains may react or link together to form a non-reactive
molecule, as follows:
thus terminating the growth of each chain or an active chain end may react with an
initiator or other chemical species having a single active bond, as follows:

with the resultant cessation of chain growth. Polyethylene, polypropylene, PVC, and
polystyrene are synthesized using addition polymerization.

Condensation polymerization, also known as step growth polymerization, involves more

than one monomer species; and there is usually a small molecular weight by-product such
as water, which is eliminated. The repeat unit here forms from original monomers, and no
product has the chemical formula of mere one mer repeat unit. The polymerization of
dimethyl terephthalate and ethylene glycol to produce polyester is an important example.
The by-product, methyl alcohol, is condensed off and the two monomers combine to
produce a larger molecule (mer repeat unit). Another example, consider the formation of
a polyester from the reaction between ethylene glycol and adipic acid; the intermolecular
reaction is as follows:

This stepwise process is successively repeated, producing, in this case, a linear molecule.
The intermolecular reaction occurs every time a mer repeat unit is formed. Reaction times
for condensation are generally longer than for addition polymerization. Polyesters,
phenol-formaldehyde, nylons, polycarbonates etc are produced by condensation
polymerization. Condensation polymerization reactions also occur in sol-gel processing
of ceramic materials. Some polymers such as nylon may be polymerized by either
technique.

Polymers, unlike organic/inorganic compounds, do not have a fixed molecular weight. It

is specified in terms of degree of polymerization number of repeat units in the chain or
ration of average molecular weight of polymer to molecular weight of repeat unit.
Average molecular weight is however defined in two ways. Weight average molecular
weight is obtained by dividing the chains into size ranges and determining the fraction of
chains having molecular weights within that range. Number average molecular weight is
based on the number fraction, rather than the weight fraction, of the chains within each
size range. It is always smaller than the weight average molecular weight.

Most of polymer properties are intrinsic i.e. characteristic of a specific polymer. Foreign
substances called additives are intentionally introduced to enhance or modify these
properties. These include fillers, plasticizers, stabilizers, colorants, and flame
retardants. Fillers are used to improve tensile and compressive strength, abrasion
resistance, dimensional stability etc. wood flour, sand, clay, talc etc are example for
fillers. Plasticizers aid in improving flexibility, ductility and toughness of polymers by
lowering glass transition temperature of a polymer. These are generally liquids of low
molecular weight. Stabilizers are additives which counteract deteriorative processes such
as oxidation, radiation, and environmental deterioration. Colorants impart a specific color
to a polymer, added in form of either dyes (dissolves) or pigments (remains as a separate
phase). Flame retardants are used to enhance flammability resistance of combustible
polymers. They serve the purpose by interfering with the combustion through the gas
phase or chemical reaction.

Polymeric materials are formed by quite many different techniques depending on (a)
whether the material is thermoplast or thermoset, (b) melting/degradation temperature, (c)
atmospheric stability, and (d) shape and intricacy of the product. Polymers are often
formed at elevated temperatures under pressure. Thermoplasts are formed above their
glass transition temperatures while applied pressure ensures that the product retain its
shape. Thermosets are formed in two stages making liquid polymer, then molding it.

Different molding techniques are employed in fabrication of polymers. Compression

molding involves placing appropriate amount of polymer with additives between heated
male and female mold parts. After pouring polymer, mold is closed, and heat and
pressure are applied, causing viscous plastic to attain the mold shape. Figure-11.1 shows
a typical mould employed for compression molding.

Figure 11.1: Schematic diagram of a mould employed for compression molding

Transfer molding differs from compression molding in how the materials is introduced
into the mold cavities. In transfer molding the plastic resin is not fed directly into the
mold cavity but into a chamber outside the mold cavities. When the mold is closed, a
plunger forces the plastic resin into the mold cavities, where and molded material cures.
In injection molding, palletized materials is fed with use of hopper into a cylinder where
charge is pushed towards heating chamber where plastic material melts, and then molten
plastic is impelled through nozzle into the enclosed mold cavity where product attains its
shape. Most outstanding characteristic of this process is the cycle time which is very
short. The schematic diagram of injection-molding machine is shown in figure-11.2

Figure 11.2: Schematic diagram of injection-molding machine

Extrusion is another kind of injection molding, in which a thermoplastic material is

forced through a die orifice, similar to the extrusion of metals. This technique is
especially adapted to produce continuous lengths with constant cross-section. The
schematic diagram of a simple extrusion machine is shown in figure- 11.3

Figure 11.3: Schematic diagram of a simple extrusion machine

Blow molding of plastics is similar to blowing of glass bottles. Polymeric materials may
be cast similar to metals and ceramics.

11.2 Crystallization, melting and glass transition

Polymers are known by their high sensitivity of mechanical and/or thermal properties.
This section explains their thermal behavior. During processing of polymers, they are
cooled with/without presence of presence from liquid state to form final product. During
cooling, an ordered solid phase may be formed having a highly random molecular
structure. This process is called crystallization. The melting occurs when a polymer is
heated. If the polymer during cooling retains amorphous or non-crystalline state i.e.
disordered molecular structure, rigid solid may be considered as frozen liquid resulting
from glass transition. Thus, enhancement of either mechanical and/or thermal properties
needs to consider crystallization, melting, and the glass transition.

Crystallization and the mechanism involved play an important role as it influences the
properties of plastics. As in solidification of metals, polymer crystallization involves
nucleation and growth. Near to solidification temperature at favorable places, nuclei
forms, and then nuclei grow by the continued ordering and alignment of additional
molecular segments. Extent of crystallization is measured by volume change as there will
be a considerable change in volume during solidification of a polymer. Crystallization
rate is dependent on crystallization temperature and also on the molecular weight of the
polymer. Crystallization rate decreases with increasing molecular weight.

Melting of polymer involves transformation of solid polymer to viscous liquid upon

heating at melting temperature, Tm. Polymer melting is distinctive from that of metals in
many respects melting takes place over a temperature range; melting behavior depends
on history of the polymer; melting behavior is a function of rate of heating, where
increasing rate results in an elevation of melting temperature. During melting there
occurs rearrangement of the molecules from ordered state to disordered state. This is
influenced by molecular chemistry and structure (degree of branching) along with chain
stiffness and molecular weight.

Glass transition occurs in amorphous and semi-crystalline polymers. Upon cooling, this
transformation corresponds to gradual change of liquid to rubbery material, and then rigid
solid. The temperature range at which the transition from rubbery to rigid state occurs is
termed as glass transition temperature, Tg. This temperature has its significance as abrupt
changes in other physical properties occur at this temperature. Glass transition
temperature is also influenced by molecular weight, with increase of which glass
transition temperature increases. Degree of cross-linking also influences the glass
transition such that polymers with very high degree of cross-linking do not experience a
glass transition. The glass transition temperature is typically 0.5 to 0.75 times the
absolute melting temperature. Above the glass transition, non-crystalline polymers show
viscous behavior, and below the glass transition they show glass-brittle behavior (as
chain motion is very restricted), hence the name glass transition.
Melting involves breaking of the inter-chain bonds, so the glass- and melting-
temperatures depend on:

chain stiffness (e.g., single vs. double bonds)

size, shape of side groups
size of molecule
side branches, defects
cross-linking

11.3 Mechanical behavior of polymers

Polymer mechanical properties can be specified with many of the same parameters that
are used for metals such as modulus of elasticity, tensile/impact/fatigue strengths, etc.
However, polymers are, in many respects, mechanically dissimilar to metals. To a much
greater extent than either metals or ceramics, both thermal and mechanical properties of
polymers show a marked dependence on parameters namely temperature, strain rate, and
morphology. In addition, molecular weight and temperature relative to the glass transition
play an important role that are absent for other type of materials.

A simple stress- strain curve can describe different mechanical behavior of various
polymers. As shown in figure 11.4, the stress-strain behavior can be brittle, plastic and
highly elastic (elastomeric or rubber-like). Mechanical properties of polymers change
dramatically with temperature, going from glass-like brittle behavior at low temperatures
to a rubber-like behavior at high temperatures. Highly crystalline polymers behave in a
brittle manner, whereas amorphous polymers can exhibit plastic deformation. These
phenomena are highly temperature dependent, even more so with polymers than they are
with metals and ceramics. Due to unique structures of cross-linked polymers, recoverable
deformations up to very high strains / point of rupture are also observed with polymers
(elastomers). Tensile modulus (modulus) and tensile strengths are orders of magnitude
smaller than those of metals, but elongation can be up to 1000 % in some cases. The
tensile strength is defined at the fracture point and can be lower than the yield strength.
 Figure-11.4: Typical stress-strain curves for polymers.

As the temperature increases, both the rigidity and the yield strength decrease, while the
elongation increases. Thus, if high rigidity and toughness are the requirements, the
temperature consideration is important. In general, decreasing the strain rate has the same
influence on the strain-strength characteristics as increasing the temperature: the material
becomes softer and more ductile. Despite the similarities in yield behavior with
temperature and strain rate between polymers, metals, and ceramics, the mechanisms are
quite different. Specifically, the necking of polymers is affected by two physical factors
that are not significant in metals: dissipation of mechanical energy as heat, causing
softening magnitude of which increases with strain rate; deformation resistance of the
neck, resulting in strain-rate dependence of yield strength. The relative importance of
these two factors depends on materials, specimen dimensions and strain rate. The effect
of temperature relative to the glass transition is depicted in terms of decline in modulus
values. Shallow decline of modulus is attributed to thermal expansion, whereas abrupt
changes are attributable to viscoelastic relaxation processes.

Together molecular weight and crystallinity influence a great number of mechanical

properties of polymers including hardness, fatigue resistance, elongation at neck, and
even impact strength. The chance of brittle failure is reduced by raising molecular
weight, which increases brittle strength, and by reducing crystallinity. As the degree of
crystallinity decreases with temperature close to melting point, stiffness, hardness and
yield strength decrease. These factors often set limits on the temperature at which a
polymer is useful for mechanical purposes.

Elastomers, however, exhibit some unique mechanical behavior when compared to

conventional plastics. The most notable characteristics are the low modulus and high
deformations as elastomers exhibit large, reversible elongations under small applied
stresses. Elastomers exhibit this behavior due to their unique, cross-linked structure.
Elastic modulus of elastomers (resistance to the uncoiling of randomly orientated chains)
increases as with increase in temperature. Unlike non-cross-linked polymers, elastomers
exhibit an increase inelastic modulus with cross-link density.

11.4 Mechanisms of deformation and strengthening of polymers

An understanding of deformation mechanisms of polymers is important in order to be

able to manage the optimal use of these materials, a class of materials that continues to
grow in terms of use in structural applications. Despite the similarities in ductile and
brittle behavior with to metals and ceramics respectively, elastic and plastic deformation
mechanisms in polymers are quite different. This is mainly due to (a) difference in
structure they made of and (b) size of the entities responsible for deformation. Plastic
deformation in metals and ceramics can be described in terms of dislocations and slip
planes, whereas polymer chains must undergo deformation in polymers leading to
different mechanism of permanent deformation. Unique to most of the polymers is the
viscoelasticity means when an external force is applied, both elastic and plastic
(viscous) deformation occur. For viscoelastic materials, the rate of strain determines
whether the deformation in elastic or viscous. The viscoelastic behavior of polymeric
materials is dependent on both time and temperature.

Plastic polymers deform elastically by elongation of the chain molecules from their stable
conformations in the direction of the applied stress by the bending and stretching of the
strong covalent bonds. In addition, there is a possibility for slight displacement of
adjacent molecules, which is resisted by weak secondary / van der Waals bonds. Plastic
deformation in polymers is not a consequence of dislocation movement as in metals.
Instead, chains rotate, stretch, slide and disentangle under load to cause permanent
deformation. This permanent deformation in polymers might occur in several stages of
interaction between lamellar and intervening amorphous regions. Initial stages involve
elongation of amorphous tie chains, and eventual alignment in the loading direction.
Continues deformation in second stage occurs by the tilting of the lamellar blocks. Next,
crystalline block segments separate before blocks and tie chains become orientated in the
direction of tensile axis in final stage. This leads to highly orientated structure in
deformed polymers.

Elastomers, on the other hand, deform elastically by simple uncoiling, and straightening
of molecular chains that are highly twisted, kinked, and coiled in unstressed state. The
driving force for elastic deformation is change in entropy, which is a measure of degree
of disorder in a system. When an elastomer is stretched, the systems order increases. If
elastomer is released from the applied load, its entropy increases. This entropy effect
results in a rise in temperature of an elastomer when stretched. It also causes the modulus
of elasticity to increase with increasing temperature, which is opposite to the behavior of
other materials.

Fracture of polymers is again dependent on morphology of a polymer. As a thumb rule,

thermosets fracture in brittle mode. It involves formation of cracks at regions where there
is a localized stress concentration. Covalent bonds are severed during the fracture.
However, both ductile and brittle modes are possible mode of fracture for thermoplasts.
Many of thermoplasts can exhibit ductile-to-brittle transition assisted by reduction in
temperature, increase in strain rate, presence of notch, increased specimen thickness and
a modification of the polymer structure. Unique to polymer fracture is crazing presence
of regions of very localized yielding, which lead to formation of small and interconnected
microvoids. Crazes form at highly stressed regions associated with scratches, flaws and
molecular inhomogeneties; and they propagate perpendicular to the applied tensile stress
and typically are 5 m or less thick. A craze is different from a crack as it can support a
load across its face.

The deformation of plastic materials can be primarily elastic, plastic, or a combination of

both types. The deformation mode and resistance of deformation depends on many
parameters for different plastics. The following factors influence the strength of a
thermoplast: average molecular mass, degree of crystallization, presence of side groups,
presence of polar and other specific atoms, presence of phenyl rings in main chains and
addition of reinforcements. Effect of every one of these factor can be used to strengthen a
thermoplast. Thermosets are, however, strengthened by reinforcement methods.

Strength of a thermoplast is directly dependent on its average molecular mass since

polymerization up to a certain molecular-mass range is necessary to produce a stable
solid. This method is not used so often as after a critical mass range, increasing the
average molecular mass does not greatly increase its strength. In general, as the degree of
crystallinity increases, the strength, modulus and density all increase for a thermoplast.
Another method to increase the strength is to create more resistance to chain slippage.
This can be achieved by addition of bulky side groups on main chains, which results in
increase of strength but reduces the ductility. Increased resistance to chain slippage can
be achieved by increasing the molecular bonding forces between the polymer chains.
E.g.: introducing a chlorine atom on every other carbon atom of main chain to make
polyvinylchloride (PVC). Introducing an ether linkage (i.e. introduction of oxygen atom)
or amide linkage (i.e. introduction of oxygen and nitrogen atoms) into the main chain can
increase the rigidity of thermoplasts. One of the most important strengthening methods
for thermoplasts is the introduction of phenylene rings in the main chain. It is commonly
used for high-strength engineering plastics. The phenylene rings cause steric hindrance to
rotation within the polymer chain and electronic attraction of resonating electrons
between adjacent molecules. Another method of strengthening is introduction of
reinforcements like glass fibers. Glass content ranges from 20 to 40%, depending on
trade-off between desired strength, ease of processing and economics.

Thermosets are strengthened by reinforcements again. Different reinforcements are in use

according to the necessity. Glass fibers are most commonly used to form structural and
molding plastic compounds. Two most important types of glass fibers are E (electrical)-
and S (high strength)- glasses. E-glass (lime-aluminium-borosilicate glass with zero or
low sodium and potassium levels) is often used for continuous fibers. S-glass (65%SiO2,
25%Al2O3 and 10% MgO) has higher strength-to-weight ratio and is more expansive thus
primary applications include military and aerospace applications. Carbon fiber reinforced
plastics are also often used in aerospace applications. However they are very expansive.
The other classes of reinforcements include aramid (aromatic polyamide) fibers. They are
popularly known as Kevlar. Presently two commercial variants of Kevlar are available
Kevlar29 and Kevlar49. Kevlar29 is a low-density, high strength aramid fiber designed
for applications such as ballistic protection, ropes and cables. Kevlar49 is characterized
by a low density and high strength/modulus; is used in aerospace, marine, automotive and
other industrial applications. Thermosets without reinforcements are strengthened by
creation of network of covalent bonds throughout the structure of the material. Covalent
bonds can be developed during casting or pressing under heat and pressure.

11.5 Characteristics and typical applications of few plastic materials.

a) Thermo plastics

1. Acrylonitrile-butadiene-styrene (ABS):
Characteristics: Outstanding strength and toughness, resistance to heat distortion; good
electrical properties; flammable and soluble in some organic solvents.
Application: Refrigerator lining, lawn and garden equipment, toys, highway safety
devices.

2. Acrylics (poly-methyl-methacrylate)
Characteristics: Outstanding light transmission and resistance to weathering; only fair
mechanical properties.
Application: Lenses, transparent aircraft enclosures, drafting equipment, outdoor signs

3. Fluorocarbons (PTFE or TFE)

Characteristics: Chemically inert in almost all environments, excellent electrical
properties; low coefficient of friction; may be used to 260o C; relatively weak and poor
cold-flow properties.
Application: Anticorrosive seals, chemical pipes and valves, bearings, anti adhesive
coatings, high temperature electronic parts.

4. Polyamides (nylons)
Characteristics: Good mechanical strength, abrasion resistance, and toughness; low
coefficient of friction; absorbs water and some other liquids.
Application: Bearings, gears, cams, bushings, handles, and jacketing for wires and cables

5. Polycarbonates
Characteristics: Dimensionally stable: low water absorption; transparent; very good
impact resistance and ductility.
Application: Safety helmets, lenses light globes, base for photographic film

6. Polyethylene
Characteristics: Chemically resistant and electrically insulating; tough and relatively low
coefficient of friction; low strength and poor resistance to weathering.
Application: Flexible bottles, toys, tumblers, battery parts, ice trays, film wrapping
materials.
7. Polypropylene
Characteristics: Resistant to heat distortion; excellent electrical properties and fatigue
strength; chemically inert; relatively inexpensive; poor resistance to UV light.
Application: Sterilizable bottles, packaging film, TV cabinets, luggage

8. Polystyrene
Characteristics: Excellent electrical properties and optical clarity; good thermal and
dimensional stability; relatively inexpensive
Application: Wall tile, battery cases, toys, indoor lighting panels, appliance housings.

9. Polyester (PET or PETE)

Characteristics: One of the toughest of plastic films; excellent fatigue and tear strength,
and resistance to humidity acids, greases, oils and solvents
Application: Magnetic recording tapes, clothing, automotive tire cords, beverage
containers.

b) Thermo setting polymers

1. Epoxies
Characteristics: Excellent combination of mechanical properties and corrosion
resistance; dimensionally stable; good adhesion; relatively inexpensive; good electrical
properties.
Application: Electrical moldings, sinks, adhesives, protective coatings, used with
fiberglass laminates.

2. Phenolics
Characteristics: Excellent thermal stability to over 150o C; may be compounded with a
large number of resins, fillers, etc.; inexpensive.
Application: Motor housing, telephones, auto distributors, electrical fixtures.

References

1. V. R. Gowariker, N. V. Viswanathan, and Jayadev Sreedhar, Polymer Science,

New Age International (P) Limited publishers, Bangalore, 2001
2. C. A. Harper, Handbook of Plastics Elastomers and Composites, Third Edition,
McGrawHill Professional Book Group, New York, 1996.
3. William D. Callister, Jr, Materials Science and Engineering An introduction,
sixth edition, John Wiley & Sons, Inc. 2004.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 12. Composites

There is a great need for materials with special properties with emergence of new
technologies. However, conventional engineering materials are unable to meet this
requirement of special properties like high strength and low density materials for aircraft
applications. Thus, emerged new class of engineering materials composites.
Unfortunately, there is no widely accepted definition for a composite material. For the
purpose of this module, the following definition is adopted: any multiphase material that
is artificially made and exhibits a significant proportion of the properties of the
constituent phases. The constituent phases of a composite are usually of macro sized
portions, differ in form and chemical composition and essentially insoluble in each other.

Composites are, thus, made by combining two distinct engineering materials in most
cases; one is called matrix that is continuous and surrounds the other phase dispersed
phase. The properties of composites are a function of the properties of the constituent
phases, their relative amounts, and size-and-shape of dispersed phase.

Millions of combinations of materials are possible and thus so number of composite

materials. For ease of recognition, composite materials are classified based on different
criteria like: (1) type of matrix material metal matrix composites, polymer matrix
composites and ceramic matrix composites (2) size-and-shape of dispersed phase
particle-reinforced composites, fiber-reinforced composites and structural composites. It
is understandable that properties of composite materials are nothing but improved version
of properties of matrix materials due to presence of dispersed phase. However, engineers
need to understand the mechanics involved in achieving the better properties. Hence the
following sections highlight the mechanics of composites, which depend on size-and-
shape of dispersed phase.

12.1 Particle-reinforced composites

This class of composites is most widely used composites mainly because they are widely
available and cheap. They are again two kinds: dispersion-strengthened and particulate-
reinforced composites. These two classes are distinguishable based upon strengthening
mechanism dispersion-strengthened composites and particulate composites.

In dispersion-strengthened composites, particles are comparatively smaller, and are of

0.01-0.1m in size. Here the strengthening occurs at atomic/molecular level i.e.
mechanism of strengthening is similar to that for precipitation hardening in metals where
matrix bears the major portion of an applied load, while dispersoids hinder/impede the
motion of dislocations. Examples: thoria (ThO2) dispersed Ni-alloys (TD Ni-alloys) with
high-temperature strength; SAP (sintered aluminium powder) where aluminium matrix
is dispersed with extremely small flakes of alumina (Al2O3).

Figure 12.1. Particulate reinforced composite

Figure 12.2. Particulate reinforced composite

Particulate composites are other class of particle-reinforced composites. These contain

large amounts of comparatively coarse particles. These composites are designed to
produce unusual combinations of properties rather than to improve the strength.
Mechanical properties, such as elastic modulus, of particulate composites achievable are
in the range defined by rule of mixtures as follows:

Upper bound is represented by: E c (u ) = E mVm + E pV p

Em E p
And lower bound is represented by: E c (l ) =
E pVm + E mV p

where E and V denote elastic modulus and volume fractions respectively while c, m, and
p represent composite, matrix and particulate phases. A schematic diagram of these
bounds is shown in the figure-12.1.
 Figure-12.3: Schematic presentation of rule-of-mixture bounds.

Particulate composites are used with all three material types metals, polymers and
ceramics. Cermets contain hard ceramic particles dispersed in a metallic matrix. Eg.:
tungsten carbide (WC) or titanium carbide (TiC) embedded cobalt or nickel used to make
cutting tools. Polymers are frequently reinforced with various particulate materials such
as carbon black. When added to vulcanized rubber, carbon black enhances toughness and
abrasion resistance of the rubber. Aluminium alloy castings containing dispersed SiC
particles are widely used for automotive applications including pistons and brake
applications.

Concrete is most commonly used particulate composite. It consists of cement as binding

medium and finely dispersed particulates of gravel in addition to fine aggregate (sand)
and water. It is also known as Portland cement concrete. Its strength can be increased by
additional reinforcement such as steel rods/mesh.

12.2 Fiber-reinforced composites

Most fiber-reinforced composites provide improved strength and other mechanical

properties and strength-to-weight ratio by incorporating strong, stiff but brittle fibers into
a softer, more ductile matrix. The matrix material acts as a medium to transfer the load to
the fibers, which carry most off the applied load. The matrix also provides protection to
fibers from external loads and atmosphere.

These composites are classified as either continuous or discontinuous. Generally, the

highest strength and stiffness are obtained with continuous reinforcement. Discontinuous
fibers are used only when manufacturing economics dictate the use of a process where
the fibers must be in this form.
The mechanical properties of fiber-reinforced composites depend not only on the
properties of the fiber but also on the degree of which an applied load is transmitted to
the fibers by the matrix phase. Length of fibers, their orientation and volume fraction in
addition to direction of external load application affects the mechanical properties of
these composites.

Effect of fiber length: Some critical length (lc) is necessary for effective strengthening and
stiffening of the composite material, which is defined as:

*f d
lc =
2 c

*f ultimate/tensile strength of the fiber, d diameter of the fiber, c interface bond

strength. Fibers for which l >> lc (normally l >15 lc) are termed as continuous,
discontinuous or short fibers on the other hand.

Effect of fiber orientation and concentration: with respect to orientation, two extremes
possibilities are parallel alignment and random alignment. Continuous fibers are
normally aligned, whereas discontinuous fibers are randomly or partially orientated. Two
instants of loading are: longitudinal loading and transverse loading.

(a) Continuous fiber composites:

Under longitudinal loading, by assuming that deformation of both matrix and fiber is the
same i.e. isostrain condition, rule-of-mixtures results in the following:

Am Af
c =m + f
Ac Ac

where Am/Ac and Af/Ac are the area fractions of the matrix and fiber phases respectively.
In the composite, if matrix and fiber are all of equal length, area fractions will be equal to
volume fractions. Thus,

c = mVm + f V f

When the isostrain assumption is taken into account, the above equation transforms into

E cl = E mVm + E f V f = E m (1 V f ) + E f V f

The ratio of the load carried by the fibers to that carried by the matrix is given by

Ff E fVf
=
Fm E mVm
In case of transverse loading, it is assumed that both matrix and fiber will experience the
equal stress i.e. isostress condition. Then the modulus of the composite is given by:

Em E f Em E f
E ct = =
E f Vm + E mV f E f (1 V f ) + E mV f

Longitudinal tensile strength: as mentioned earlier, matrix material is softer i.e. fibers
strain less and fail before the matrix. And once the fibers have fractured, majority of the
load that was borne by fibers is now transferred to the matrix. Based on this criterion the
following equation can be developed for longitudinal strength of the composite:

cl* = m' (1 V f ) + *f V f

where m stress in the matrix at fiber failure, *f fiber tensile strength.

Whereas longitudinal strength is dominated by fiber strength, a variety of factors will

have a significant influence on the transverse strength. These include properties of both
the fiber and matrix, interface bond strength, and the presence of voids.

(b) Discontinuous and aligned fiber composites:

Even though reinforcement efficiency is lower for discontinuous fiber composites than
continuous fiber composites, discontinuous and aligned fiber composites are
commercially gaining an important place. The longitudinal strength of these composites
is given by:

lc
cd* = *f V f (1 ) + m' (1 V f ) when l > lc and
2l

l c
cd* ' = V f + m' (1 V f ) when l < lc
d

where c smaller of either the fiber-matrix bond strength or the matrix shear yield
strength.

(c) Discontinuous and randomly orientated fiber composites:

Reinforcement efficiency of these fiber composites is difficult to calculate, and is usually

characterized by a parameter known as fiber efficiency parameter, K. K depends on Vf
and the Ef/Em ratio. If rule-of-mixtures can be applied, elastic modulus of these
composites is given by:

E cl = K ( E mVm + E f V f )
12. 3 Structural composites

These are special class of composites, usually consists of both homogeneous and
composite materials. Properties of these composites depend not only on the properties of
the constituents but also on geometrical design of various structural elements. Two
classes of these composites widely used are: laminar composites and sandwich structures.

Laminar composites: there are composed of two-dimensional sheets/layers that have a

preferred strength direction. These layers are stacked and cemented together according to
the requirement. Materials used in their fabrication include: metal sheets, cotton, paper,
woven glass fibers embedded in plastic matrix, etc. Examples: thin coatings, thicker
protective coatings, claddings, bimetallics, laminates. Many laminar composites are
designed to increase corrosion resistance while retaining low cost, high strength or light
weight.

Sandwich structures: these consist of thin layers of a facing material joined to a light
weight filler material. Neither the filler material nor the facing material is strong or rigid,
but the composite possesses both properties. Example: corrugated cardboard. The faces
bear most of the in-plane loading and also any transverse bending stresses. Typical face
materials include Al-alloys, fiber-reinforced plastics, titanium, steel and plywood. The
core serves two functions it separates the faces and resists deformations perpendicular
to the face plane; provides a certain degree of shear rigidity along planes that are
perpendicular to the faces. Typical materials for core are: foamed polymers, synthetic
rubbers, inorganic cements, balsa wood. Sandwich structures are found in many
applications like roofs, floors, walls of buildings, and in aircraft for wings, fuselage and
tailplane skins.

References

1. K. K. Chawla, Composite Materials Science and Engineering, Second Edition,

Springer-Verlag, New York, 1998.
2. William D. Callister, Jr, Materials Science and Engineering An introduction,
sixth edition, John Wiley & Sons, Inc. 2004.
3. D. Hull and T. W. Clyne, An Introduction to Composite Materials, Second
Edition, Cambridge University Press, New York, 1996.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 13. Corrosion and Degradation of Materials

All engineering materials are subjected to numerous external mechanical and

environmental factors during their service. Those factors include temperature, chemical
attack, mechanical vibration, applied mechanical loads, etc. Under the influence of these
factors, engineering materials gets degraded i.e. they loss their potential to perform the
task. The degradation is inevitable, and may lead to catastrophes! Hence, degradation of
materials not only be monitored but also be factored into the design of different
components by engineers.

Mechanism of degradation is different for each engineering material group. Material loss
either by dissolution or by formation of nonmetallic scale/film occurs in metals, while
ceramics are relatively resistant to degradation. However, ceramics also gets deteriorated
at extreme temperatures. Polymers, on the other hand, may dissolve/distort in presence of
a liquid solvent or even when exposed to electromagnetic radiation. Degradation due to
chemical/electro-chemical factors is termed as corrosion, while oxidation refers to
formation of nonmetallic scales in metals.

13.1 Corrosion of Metals

Corrosion can be defined as unintentional deterioration of materials due to

electrochemical factors. It usually starts at the external material surface. The corrosion
resistance of metals and alloys is a basic property related to the easiness with which these
materials react with a given environment. Corrosion is a natural process that seeks to
reduce the binding energy in metals, and is of electrochemical in nature. The end result of
corrosion involves a metal atom being oxidized, whereby it loses one or more electrons
and leaves the bulk metal. The process is called oxidation, some times also called anodic
reaction. The lost electrons are conducted through the bulk metal to another site where
they are reduced, the process called reduction. The site where metal atoms lose electrons
is called the anode, and the site where electrons are transferred to the reducing species is
called the cathode. As a consequence of these electrochemical reactions, metal ions either
go into the corroding solution as ions, or they may form an insoluble compound with
nonmetallic elements.
Corrosion tendencies for different metals in measured in terms of electromotive force
(EMF). A series of metals are arranged according to their potential to get corroded with
respect to hydrogen electrode. Table13.1 presents EMF series, and those at top are noble,
while those metals at bottom are more prone to corrosion than the metals above them in
the series.

As corrosion is electrochemical in nature, temperature and concentration of chemical

does have an important role to play. In general, with increase of either or both
temperature and concentration, corrosion rate increases. The electrode potential depends
on the concentration of the electrolyte. At 25C, the following equation, known as Nernst
equation, gives the electrode potential:

0.0592
E = E0 + log(C ion )
n

where E is the electrode potential in a solution containing a concentration Cion (in moles)
of metals, n is the valence of the metallic ion, E0 is the standard electrode potential in a 1-
M solution.

Faradays equation gives the amount (w, weight in gms.) of metal plated on the cathode,
or removed from the metal by corrosion:

ItM
w=
nF

where I is the current (A), M is the atomic mass of the metal, n is the valence of the metal
ion, t is the time (secs), and F is the Faradays constant (96,500 C).

Table 13.1: Standard EMF series of metals.

Metal Electrode potential, E0 (V)

Au Au3++3e- +1.420
2H2O O2+4H++4e- +1.229
Pt Pt2++2e- +1.200
Increasingly inert Ag Ag++e- +0.800
Fe2+ Fe3++e- +0.771
4(OH)- O2+2H2O+4e- +0.401
Cu Cu2++2e- +0.340
H2 2H++2e- 0.000
Pb Pb2++2e- -0.126
Sn Sn2++2e- -0.136
Ni Ni2++2e- -0.250
Co Co2++2e- -0.277
Cd Cd2++2e- -0.403
Fe Fe2++2e- -0.440
 Cr Cr3++3e- -0.744
Zn Zn2++2e- -0.763
Al Al3++3e- -1.662
Mg Mg2++2e- -2.363
Increasingly active Na Na++e- -2.714
K K++e- -2.924
Li Li++e- -3.050

Galvanic Series
Anodic or Least Noble end
Magnesium
Magnesium Alloys
Zinc
Aluminum 5052, 3004, 3003, 1100, 6053
Cadmium
Aluminum 2117, 2017, 2024
Mild Steel (1018), Wrought Iron
Cast Iron, Low Alloy High Strength Steel
Chrome Iron (Active)
Stainless Steel, 430 Series (Active)
302, 303, 321, 347, 410,416, Stainless Steel (Active)
Ni - Resist
316, 317, Stainless Steel (Active)
Carpenter 20cb-3 Stainless (Active)
Aluminum Bronze (Ca 687)
Hastelloy C (Active) Inconel 625 (Active) Titanium (Active)
Lead - Tin Solders
Lead
Tin
Inconel 600 (Active)
Nickel (Active)
60 Ni-15 Cr (Active)
80 Ni-20 Cr (Active)
Hastelloy B (Active)
Brasses
Copper (Ca102)
Manganese Bronze (Ca 675), Tin Bronze (Ca903, 905)
Silicone Bronze
Nickel Silver
Copper - Nickel Alloy 90-10
Copper - Nickel Alloy 80-20
430 Stainless Steel
Nickel, Aluminum, Bronze (Ca 630, 632)
Monel 400, K500
Silver Solder
Nickel (Passive)
60 Ni- 15 Cr (Passive)
Inconel 600 (Passive)
80 Ni- 20 Cr (Passive)
Chrome Iron (Passive)
302, 303, 304, 321, 347, Stainless Steel (Passive)
316, 317, Stainless Steel (Passive)
Carpenter 20 Cb-3 Stainless (Passive), Incoloy 825nickel - Molybdeum - Chromium -
Iron Alloy (Passive)
Silver
Titanium (Pass.) Hastelloy C & C276 (Passive), Inconel 625(Pass.)
Graphite
Zirconium
Gold
Platinum
Cathode Or Most Noble end

Passivity: Under particular environments, some active metals and alloys may turn inert,
known as passivity. It is usually observed in metals such as Cr, Fe, Ni, Ti and their alloys.
Passivity is inherited as a result of formation of a thin and highly adherent oxide film on
the metal surface. For example: stainless steels are highly resistant to corrosion in many
environments, except in hard sea water. For steels, Cr forms a protective layer.
Aluminium is also observed to passive in many environments. This is due to formation
thin oxide layer.

Corrosion in metals is classified based on the manner in which it is manifest into eight
forms, namely: uniform, galvanic, crevice, pitting, inter-granular, selective leaching,
erosion-corrosion, and stress corrosion.

Uniform corrosion: As the name suggests, electrochemical corrosion occurs over entire
exposed surface with equal intensity. It usually results in formation of a scale/film. This
form of corrosion is most common, easy to monitor, and less damaging than the other
forms. Painting the surface is best counter measure for it.

Galvanic corrosion: This occurs when two metals with different compositions i.e. EMF
are electrically connected in presence of an electrolyte. Out-of-these two metals, less inert
metal will corrode, while the other will be protected from the corrosion. Corrosion rate
depends on the relative anode-to-cathode surface areas that are exposed to the electrolyte.
the following measures are observed to reduce the damage due to this form of corrosion:
when coupling of two different metals is unavoidable, choose that are close together in
EMF series; tailor the anode-to-cathode area ratio; metals may be electrically insulated
form each other; or use a third metal to protect the both, known as cathodic protection.

Crevice corrosion: This occurs as a result of concentration difference between two

regions of same metal component. Corrosion takes place in the locale that is of lower
concentration. Probable sites for this corrosion are: crevices and recesses, under scales of
dirt/corrosion product. This form of corrosion may be avoided if enough care is taken to
avoid concentration differences. E.g.: welding instead of riveting. Other measures include
removing accumulated deposits; avoiding stagnant areas.

Pitting: It is another form of localized corrosion. As a consequence of this, small pits, and
holes form, hence the name. This occurs in normal direction to the surface exposed, and
so is difficult to monitor. Thus, it is extremely dangerous form of corrosion, as material
loss is often undetectable till the component fails. Same measures as for crevice corrosion
are part of care against this form of corrosion. In addition, polishing of surfaces could do
good.
Inter-granular corrosion: This occurs again due to concentration difference, but now
along grain boundaries, which are usually of different chemical composition compared
with rest of the grains. Thus grain boundaries are highly susceptible to corrosion. This
form of corrosion is very severe in stainless steels. These materials may be protected by
suitable heat treatment; lowering carbon content; addition of alloying elements those
readily forms of carbides.

Selective leaching: It is the preferential removal of a solid alloy by corrosion process.

Best example can be dezincification of brass. It also occurs as the loss of Ni, Sn and Cr
from Cu-alloys; Fe from cast iron; Ni from steel alloys, and Co from satellite. As a result
of this corrosion, mechanical properties of the components are impaired as the
component becomes porous. Measures of protection include change if composition;
change of environment; use of cathodic protection.

Erosion-corrosion: It is defined as acceleration of corrosion attack in a metal due to

relative motion of a corrosive fluid and a metal surface i.e. is result of combined action of
chemical attack and mechanical abrasion/wear. Practically, all metals and alloys are
prone to this corrosion form. It has more damaging influence on metals which passivate
by forming a protective film and also on soft metals like Cu and Pb. Counter measures
include: change of design; choosing different metal for the component; removal of
particulates and bubbles form the fluid to lessen the erosive effects.

Stress corrosion: This form of corrosion, also known as stress corrosion cracking (SCC),
occurs as a result of combined action of applied tensile stresses and corrosive
environment. During SCC, the metals surface is usually attacked very little while highly
localized cracks propagate through the metal section. The stresses which cause SCC can
be residual or applied. Only certain combinations of alloys and environments cause SCC.
E.g.: Stainless steels crack in chloride environments but not in ammonia-containing ones.
Best counter measure is elimination of external stress. Other measures are: to eliminate
detrimental environment; change of the alloy if neither environment nor stress level can
be changed; cathodic protection: addition of inhibitors.

Hydrogen embrittlement is more failure than a form of corrosion, but it is often results
from the hydrogen, produced from corrosion. Atomic hydrogen produced during
corrosion diffuses interstitially through crystal lattice, and interferes with dislocation
motion, leading to failure. It is similar to stress corrosion in the sense that ductile
materials experience brittle failures as a result. Counter measures to hydrogen
embrittlement include: heat treatment to reduce strength of the alloy; removal of source
of hydrogen; baking the component to drive out any dissolved hydrogen.

13.2 Corrosion of Ceramics

Ceramics, being compounds between metallic and nonmetallic elements, may be thought
of as having already been corroded. It is often said that one of the biggest advantages
which ceramics have over other engineering materials is their corrosion resistance, that is,
their chemical inertness in corrosive environments.
Corrosion is generally understood as property degradation due to environmental attack.
Ceramics corrode by simple chemical dissolution, in contrast to the electrochemical
processes involved with corrosion of metals. There are a number of environments in
which ceramics can degrade at a rapid rate. There exists a tremendous need for reliable
and corrosion resistant structural ceramic or partly ceramic materials which can be used
in aggressive environments such as:

- high energy battery systems (such as sodium-sulphur): beta-alumina is being

investigated

- gas turbines: silicon nitride and/or carbide are being investigated

- heat exchangers: SiC, composites are being investigated

Ceramics are indeed much more environmentally stable, as compared to any other group
of engineering materials, e.g. metals or plastics. Still, the potential for ceramics as
corrosion resistant engineering structural materials are far from being fully realized,
because of:

- mechanical non-reliability of structural ceramic components

- difficult design with brittle materials

- a shortage of information and standardization of ceramics

- human reluctance to use non-ductile materials

13.3 Degradation of polymers

While the plastics industry searches for solutions to the problem of plastics waste, there
is, surprisingly, a growing band of people trying to save plastics. The crucial fact is that
plastics are organic and can are prone to physiochemical attacks. Polymeric materials
degrade by non-corrosive processes.

As all other engineering materials, plastics do indeed degrade during their service.
However, degradation of plastics is not termed as corrosion, as it is of physiochemical in
nature in contrast to electrochemical corrosion of metals. Degradation of polymers, thus,
involves a wide variety of reactions and results like absorption and swelling, dissolution,
bond rupture due to heat, chemical effects, or radiation; weathering, etc. Degradation of
polymers might be due to exposure to light (especially UV), humidity, oxygen, heat,
bacteria or external loads/stress. However, due to complex nature of chemical bonds in
polymers, their degradation mechanisms are not well understood.

Swelling and dissolution: Polymers, when exposed to liquids, they get swelling as a result
of solute diffusion and absorption of solute. Because of swelling i.e. separation of chains,
secondary bonds become weaker. As a consequence, the material becomes softer and
more ductile. Swelling is considered as partial dissolution, while the dissolution or
complete solubility is worse than the swelling. Swelling and dissolution effects are
influenced by temperature because of their physiochemical nature.

Bond rupture: Bond rupture in polymers due to degradation is known as scission.

Polymers bonds may get rupture due to many effects like radiation, heat energy, or
chemical reactions. When polymers are exposed to certain types of radiation, which may
result in broken bonds and rearrangement of atoms leads to degradation of polymers. At
elevated temperatures, bonds in polymers may get weakened, leading to deterioration of
polymers. Some chemical elements like oxygen, ozone can alter the chain scission rate as
a result of chemical reactions. This is especially pronounced in vulcanized rubbers.

Weathering: Polymers are exposed to outdoor conditions in many instants of their

service. Any degradation of polymer under these circumstances is termed as weathering,
which in fact is a combination of several different processes. It primarily involves
oxidation, which may be initiated by ultraviolet radiation from the Sun. polymers
resistance to weathering to very difficult to quantify and is very diverse.

Some examples of polymer degradation include:

- deterioration of acrylic paintings and pieces of art

- decomposition of photographic films

- decolorization of plastics pieces preserved in museums

Solutions to polymer degradation: As the recognition of polymer degradation improves,

conservation guidelines are beginning to emerge. High-tech solutions which could help in
theory are prohibitively expensive, but tailor made scavengers such as activated charcoal
or Ageless help to create a low oxygen environment. Ageless is a reactive powdered iron
and is normally used to prolong the shelf-life of dry foods by absorbing oxygen.
Epoxidised soya bean oil (ESBO), has also been tested with encouraging results as an
acid absorbing coating on degrading cellulose nitrate.

References

1. M. G. Fontana, Corrosion Engineering, Third Edition, McGraw-Hill, New

York, 1986.
2. ASM Handbook, Corrosion, Vol. 13, ASM Internation, Materials Park, OH,
1987.
3. William D. Callister, Jr, Materials Science and Engineering An introduction,
sixth edition, John Wiley & Sons, Inc. 2004.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 14. Electrical properties

Engineering materials are important in everyday life because of their versatile structural
properties. Other than these properties, they do play an important role because of their
physical properties. Prime physical properties of materials include: electrical properties;
thermal properties; magnetic properties; and optical properties. The electrical behaviors
of engineering materials are diverse, and so are their uses in electrical applications. Few
examples of electrical applications: copper and aluminium wires used in power
transmission; amorphous silicon in solar cells; LEDs; gallium-arsenide chips used in
modern day computers, cell phones, digital assistants; etc.

This chapter shall describe electrical properties and behavior of engineering materials in
electrical applications. First, basic electrical property, electrical conductivity, will be
described followed by different types of electrical conduction including semi- and super-
conductivities. Some peculiar phenomena like ferro-electricity and peizo-electricity are
explained in addition to electrical insulation.

14.1 Electrical conduction

Electrical conductivity of a material is defined in terms of ease with which a material

transmits an electrical current. Electrical current (I) is flow of electrons, and driving force
for the flow of electrons is called voltage (V). Ohms law relates these parameters as
follows:

VI

V = IR...(1)

where R is the materials resistance to flow of electrons through it.

V, I, and R respectively have units as volts, amperes, and ohms ().

Electrical resistance of a material is influenced by its geometric configuration; hence a
new parameter called electrical resistivity () is defined such as it is independent of the
geometry.

RA
= ...(2)
l

where A cross-sectional area perpendicular to the direction of the current, and l the
distance between points between which the voltage is applied. Units for are ohm-meters
(-m).

Reciprocal of the electrical resistivity, known as electrical conductivity (), is used to

express the electrical behavior of a material, which is indicative of the ease with which a
material allows of flow of electrons.

1 l
= = ....(3)
RA

Electrical conductivity has the following units: (-m)-1 or mho/meter.

Resistivity or conductivity does not depend on the dimensions of the material. These
properties are microstructure-sensitive instead, like many other intrinsic properties, for
example yield strength. Thus either of it (usually conductivity) allows us to compare
different materials. Solid engineering materials exhibit very wide range of electrical
conductivity (about 27 orders of variation). Hence the materials for electrical applications
are classified according to their electrical conductivity as: conductors, semiconductors
and insulators/dielectrics. Metals are conductors having conductivities in range of 107
(-m)-1, while semiconductors have conductivities in range from 10-6 to 104 (-m)-1, and
materials with conductivity lower than 10-10(-m)-1 are termed as insulators.

In components designed to conduct electrical energy, minimizing losses is important to

conserve energy as well as minimize heating. The electric power (P) lost when a current
flows through a material is given by:

P = VI = I2R..(4)

Electrical losses are also known as Joule heating losses.

Second form of Ohms law can be obtained by combining the equations (3) and (4) to
give:

I V
= J = E (5)
A l

where J current density (A/cm2), E electric field strength (V/cm).

Current density can be defined in different terms as

J = n.q.v ..(6)

where n number of charge carriers (1/cm3), q charge on each carrier (1.6x10-19C), v -

average drift velocity (cm/s), which occurs as a result of an applied electric or magnetic
field.

The term v /E is called mobility, (cm2/V.s) of the carriers, thus

= n.q. .(7)

From the above equation, the following can be concluded: electrical conductivity of a
material can be controlled by (i) controlling number of charge carriers, n or (ii)
controlling the mobility of the carriers, . Mobility is important for metals or electrical
conductors, whereas number of carriers is important for semi-conductors and insulators.

Electrons are charge carriers in metals. In ionic materials (for example ionic ceramics),
conduction is result of net movement of charged particles (cations and anions) in addition
to any electron motion. However, both electrons and holes are charge carriers in
semiconductors and in doped conductive polymers. Thus equation 7 can be written as
follows:

for metals : = ne .q e . e (8)

for semiconductors : = n.q. n + n.q. p ..(9)

for ionic materials : total = electronic + ionic .(10)

In all conductors, semiconductors, and many insulating materials, only electronic

conduction exists i.e. electrons are the only charge carriers. Ionic conduction results
because of net motion of charged ions. Movement different particles in various materials
depend on more than one parameter. These include: atomic bonding, imperfections,
microstructure, ionic compounds, diffusion rates and temperature. Because of these,
electrical conductivity of materials varies tremendously.

In metals not all electrons will accelerate in the presence of an electric field. Instead the
number of charge carriers is dependent on the arrangement of electron states/levels with
respect to energy, and the manner in which these states are occupied by electrons.
Electrons in metals are arranged into shells and sub-shells in accordance with Pauli
exclusion principle. Each of these shells / sub-shells is distinguished from every other by
their energy state/level. Basic existence of metals is dependent on metallic bond where
large groups of atoms come together and share valence electrons in what is called
metallic bond. Thus there are N atoms in the lattice with N atomic orbitals and N
molecular orbitals, many of which are of same energy. This leads to bands of electrons.
The electrons not involved in bonding remain in what is called the core band, whereas the
valence electrons that form the electron gas enter into the valence band. The remaining
un-filled orbitals form higher-energy bands, called conduction band. It is the conduction
band that gives metals and alloys the ability to freely conduct electrons.

Four different types of band structures are possible at 0 K: (1) Where one outermost band
is only partially filled with electrons. The energy corresponding to the highest filled state
at 0 K is called the Fermi energy, Ef. (2) There is overlap of an empty band and a filled
band. (3) and (4) One filled band (valence band) and an empty band (conduction band)
are separated by an energy band gap which lies between them. These are schematically
shown in the figure 14-1.

Figure 14-1: Possible electron band structures.

Band structures (1) and (2) are found in metals / conductors. Band structures (3) and (4)
are distinguished by the size of energy band gap. Narrower energy band gap i.e. size < 2
eV, is found in semiconductors, while the broader energy band gap i.e. size > 4 eV, is
found in insulators. Fermi energy for these two band structures lies within the band gap
near its center.

Metals have high conductivities because of the large number of free electrons that have
been excited into empty states above the Fermi energy. If the metal crystal lattices were
perfect and there are no lattice vibrations, the electrons would pass through the lattice un-
scattered, encountering no resistance. However, lattice vibrations and phonon scattering
play a role in disrupting the mean free path of electrons. In addition, crystalline defects
and impurity atom affect the conductivity. These scattering mechanisms act
independently on one another. Thus the effective resistivity of metals can be represented
as follows:
 = thermal + residual ...(11)

Equation 11 is some times known as Mathiessens rule. With increase of temperature,

thermal vibrations increase so the resistivity, and vice versa. In the same manner, with
increase of either defects or impurities, resistivity increases. Thus, in general, for pure
metals, the resistivity approaches zero at absolute zero temperature.

Some metals, however, such as Pb, lose all resistivity abruptly and completely at some
low temperatures. This phenomenon is called superconductivity, and the materials that
exhibit it are called superconductors. The temperature at which the resistivity vanishes is
called the critical transition temperature, Tc. Many elemental metals, solid-solution
alloys, some ceramics and intermetallic compounds exhibit superconductivity. For
example: Ti, V, Zn, W, Al, Hg, NbTi, Nb3Sn, MgB2, La-Sr-Cu oxide, YBa2Cu3O7-x,
carbon nanotubes, etc. BCS (Bardeen, Cooper and Schrieffer) theory explains that the
super conductivity is caused by electron-lattice interaction and that the superconducting
electrons consist of paired ordinary electrons called a Cooper pair.

14.2 Semiconductivity

Conductivity of semiconductors is not as high as metals / conductors. However, their

electrical properties are unique, in the sense that their electrical properties are extremely
sensitive to even minute concentrations of impurities. For intrinsic semiconductors, their
electrical behavior is based on inherent electronic structure of the pure material. On the
other hand, if the electrical properties are dominated by impurities, they are called
extrinsic semiconductors.

Intrinsic semiconduction: Energy band gap between the valence and conduction bands
can be used to classify materials as conductors, insulators, and semiconductors. The
magnitude of the band gap is characteristic of the lattice alone, and varies widely for
different crystals. In semiconductors, the valence and conduction bands do not overlap as
in metals, but they possess enough electrons in the valence band those can be promoted to
the conduction band at a certain temperature. Promotion of electrons leaves behind
positively charged holes in the valence band which maintain the charge neutrality. These
holes can be regarded as moving positive charges through the crystal with a charge
magnitude and ass as that of an electron. Conduction that arises from thermally / optically
excited electrons is called intrinsic semiconduction. It usually takes place at elevated
temperatures since sufficient thermal agitation is necessary to transfer a reasonable
number of electrons from the valence band to the conduction band. At still higher
temperatures, the concentration of thermally excited electrons in the conduction band
becomes so high that the semiconductor behaves more like a metal. The elements capable
of intrinsic conduction are limited, and include: Si, Ge, Sn, Pb, etc.
 Figure 14-2: Band structure and intrinsic conductivity.

In intrinsic semiconductors, the number of holes equals the number of mobile electrons.
The resulting conductivity can be expressed as

= ne .q e . e + nh .q h . h = ne .q e .( e + h ) ..(12)

as ne = nh, and both electron and hole possess equal magnitude of charge.

The mobility of holes is always less than those of electrons i.e. h < e. In Si and Ge, the
ratio e / h is approximately three and two respectively.

Extrinsic semiconduction: The charge carrier density can also be increased by adding
impurities of either higher or lower valence. This addition of impurities is known as
doping, and impure atoms in the element are called donor atoms.

When a pentavalent substitutional atoms like P, As, or Sb are placed into a covalently
bonded tetravalent materials such as Si or Ge, only four of their five valence electrons are
required to participate in covalent bonding, while the fifth electron remains weakly bound
to the donor atom. However, this binding energy is in order of 0.01 eV, much less than
that of a covalently bonded electron. The energy state of this electron is indicated as Ed.
Energy gap between the donor and conducting bands is much smaller than that between
the valence band and conduction band, Eg. Thermal agitation, even at room temperature,
is good enough to excite this electron to the conduction band, leaving behind a positively
charged hole in the donor band. Conductivity is due to motion of negatively charged
electrons in the conduction band, and so the material is called n-type semiconductor.
Fermi level (Ef) of n-type extrinsic semiconductor is higher than that of intrinsic
semiconductor i.e. it shifted upward in the band gap, to within the vicinity of the donor
state; and its exact position is a function of both temperature and donor concentration.
 Figure 14-3: Band structure and n-type semiconductivity.

Other possibility of susbstituion involves impurity with lower valence. For example
trivalent elements like B, Al, Ga, In, Th doped to Si or Ge. Since the donor atoms are
deficient in bonding electrons, one of their bonding orbitals will contain a hole that is
capable of accepting an electron from elsewhere in the crystal. Binding energy is small,
and promotion of an electron from the valence band to the acceptor band, Ea, leaves a
positively charged hole in the valence band that can act as a charge carrier. Materials with
this p-type extrinsic semiconduction are called p-type semiconductors. Fermi level (Ef)
for p-type semiconductors is positioned within the band gap and near to the acceptor
level.

Figure 14-4: Band structure and p-type semiconductivity.

For an n-type semiconductor, electrons in the conduction band are charge carriers and
holes participate in bonding, whereas for a p-type semiconductor, the electrons
participate in bonding, and the holes are the charge carriers. Conductivity equations for
these extrinsic semiconductors are written as:

n-type semiconductor: n.q. e .(13)

p-type semiconductor: n.q. h ..(14)

In both types of extrinsic semiconductors, the doping elements are chosen so that both
acceptor and donor levels are located closer to the corresponding energy bands and only a
small energy gap in involved when exciting electrons. Intrinsic semiconductors are pure
to the order greater than 107, while extrinsic semiconductors are made with impurity
levels in order of 1-1000 ppm.

In intrinsic semiconductors, the conductivity is dominated by the exponential relation

between temperature and band-gap. However, the conductivity of extrinsic
semiconductors is governed by competing forces: charge carrier density and charge
carrier mobility. At low temperatures, the number of charge carriers initially increases
with temperature. At intermediate temperatures, for n-type extrinsic semiconductor, most
of the donor electrons have been promoted and the number of charge carriers is nearly
independent of temperature in what is known as the exhaustion range. At higher
temperatures, the number of valence electrons excited to the conduction band greatly
exceeds the total number of electrons from the donor atoms, and extrinsic semiconductor
behaves like an intrinsic semiconductor. That is, at low temperatures the conductivity
varies with temperature as the charge carrier concentration, at higher temperatures the
charge carrier mobility dominates and the conductivity decreases with temperature, and at
still higher temperatures the extrinsic semiconductor behaves like an intrinsic
semiconductor.

14.3 Electrical conduction in ionic ceramics and in polymers

In addition to the conduction of charge via electrons, charge can be conducted via ions.
Ions are present in most crystalline ceramic materials such as oxides and halides. This
process is called ionic conduction, and may occur either in conjunction with or separately
from electronic conduction.

Several types of compounds show exceptionally high ionic conductivity. Such phases fall
into three broad categories: halide and chalcogenides of silver and copper; oxides with -
alumina structure; and oxides of fluorite structure. Conductivity of halides and
chalcogenides is result of larger number of available sites than the number of cations. In
oxides with -alumina structure (AM11O17) monovalent cation A (such as Na, K, Rb, Li)
is mobile. High dopant levels in the fluorite-type solid solutions leads to large defect
concentrations and vacancy ordering which results in rapid oxygen migrations. This is the
base for development of high-temperature superconductors. For example: La2CuO4 (Tc =
30 K), YBC compounds yttrium doped perovskite structure, YBa2Cu3O7 (Tc = 92 K).

Although ceramics are insulators, by properly engineering the point defects in ceramics,
it is possible to convert them into semiconductors. For example: Indium tin oxide (ITO)
is used as a transparent conductive coating on plate glass, and also for tough screen
displays; Ytrria-stabilized zirconia (YSZ) is used as solid electrolyte in solid oxide fuel
cells; Lithium cobalt oxide is used as solid electrolyte in lithium ion batteries.

Polymers have a band structure with a large energy gap leading to low conductivity
because their valence electrons are involved in covalent boding. Hence polymers are used
frequently in applications that require electrical insulation. In some cases however, low
conductivity is undesirable. For example; static electricity can accumulate on housing for
electronic equipment making the polymer vulnerable to electromagnetic radiation that
damages the internal devices. These problems can be solved in two approaches: (1)
introducing an additive to the polymer to improve conductivity, and (2) creating
polymers with inherent conductivity.

Adding ionic compound to the polymer can reduce resistivity because the ions migrate to
the polymer surface and attract moisture, which in turn, dissipates static charges.
Introducing conductive fillers such as carbon black can also dissipate static charges.
Some polymers inherently have good conductivity as result of doping or processing
techniques. When acetal polymers are doped with agents such as arsenic pentafluoride,
electrons or holes are able to jump freely from one atom to another along the backbone of
the chain, increasing the conductivity to near that of metals. However, unlike
semiconductors, the dopant atoms or molecules do not substitute for or replace the
polymer atoms. Other polymers include polyparaphynylene, polypyrole, and polyaniline.
Some other polymers such as polyphthaocyanine can be cross-linked by special curing
processes to raise its conductivity.

14.4 Dielectric behavior

A dielectric is a material separating two charged bodies. For a material to be a good

dielectric, it must be an electrical insulator. Thus, any insulator is also called a dielectric.
Dielectric materials are used in capacitors, devices used to store the electric energy.

When a voltage is applied between two parallel plates in a vacuum, one plate becomes
positively charged and the other negatively charged, with corresponding field directed
from the positive to the negative. Then the energy stored, Q, is given by:

Q = CV ... (15)

where C is the capacitance (units: coulombs per volt or farads [F]). It is dependent on the
geometry of plates as follows:

A
C = 0 .. (16)
l

where A area of the plates, l distance between them, and 0 permittivity of vacuum
(8.85 X 10-12 F/m). If a dielectric material is inserted between the plates,

A
C = . (17)
l

permittivity of the medium, which will be greater in magnitude than 0. The relative
permittivity r, also called dielectric constant, is defined as

r = ...(18)
0

which is greater than unity, and presents the increase in charge storing capacity by
insertion of the dielectric medium between the plates. This is result of polarization
orientation of permanent or induced dipoles under externally applied electric field.
Polarization causes positive charge to accumulate on the bottom surface next to the
negatively charged plate, and causes negative charge to accumulate toward the positively
charged plate on the top. This tends to decrease the effective surface charge density on
either plate. The expected decrease in effective charge corresponds to the polarization, of
the material which equals the induced dipole moment per unit volume of polarizable
material. The magnitude of the charge per unit area on either plate, called the dielectric
displacement or flux density or surface charge density, D, is defined as:

D = . (19)

where electric field strength,

= V / l.. (20)

In the presence of the dielectric, surface charge density can also be presented as

D = = 0 + P (21)

where polarization is defined as,

P = 0 ( r 1) ... (22)

the increase in charge density above that for a vacuum because of the presence of the
dielectric.

Figure 14-5: Schematic representation of capacitor and polarization phenomenon.

The polarization may also be thought of as the total dipole moment per unit volume of the
dielectric material or as a polarization electric field within the dielectric that results from
the mutual alignment of the many a atomic or molecular dipoles. Polarization of a
dielectric material may be caused by at least one of four major types of polarization:
electronic, ionic (atomic), orientation, and space charge (interfacial).

Electronic polarization arises because the center of the electron cloud around a nucleus is
displaced under an applied electric field. Ionic polarization occurs in ionic materials
because an applied field acts to displace cations in the direction of the applied field while
displacing anions in a direction opposite to the applied field. Orientation polarization can
occur in materials that possess permanent electric dipole moments. These permanent
dipoles tend to become aligned with the applied electric field, but entropy and thermal
effects tend to counter this alignment, thus it is highly temperature dependent. Space
charge polarization results from the accumulation of charge at structural interfaces in
heterogeneous materials. Such polarization occurs when one of the phases has a much
higher resistivity than the other.

The total polarization subjected to an alternating electric field depends upon the ease with
which the permanent or induced dipoles can reverse their alignment. The time required
for dipole reversal is called the relaxation time, and its reciprocal is called the relexation
frequency. As the frequency of the applied electric field approached the relaxation
frequency, the polarization response increasingly lags behind the applied field. The
reorientation of each type of dipole is opposed by internal friction, which leads to heating
in the sample and power loss. The power loss depends on the degree to which the
polarization lags behind the electric field. This is also called dielectric loss, and a low
dielectric loss is desired at the frequency of utilization.

At high enough frequencies, the dielectric will experience electrical breakdown initiated
by the field-induced excitation of a number of electrons into the conduction band, and the
insulator become a conductor. The magnitude of the electric field required to cause
dielectric breakdown is called the dielectric strength or breakdown strength.

Many ceramics and polymers are utilized as insulators and in capacitors. For example:
glass, porcelain, stealite, mica. These have dielectric constants within the range of 6-10.
Typical applications: electrical insulation, switch bases, light receptacles. Dielectric
constant for most polymers lies in the range of 2-5, less than that for ceramics, since the
later exhibit greater dipole moments. Typical applications: insulation for wires, cables,
motors, generators, some capacitors.

14.5 Ferro-electricity and Piezoelectricity

Two peculiar phenomena related to electric dipoles are ferro-electricity and piezo-
electricity. Ferro-electricity is defined as the spontaneous alignment of electric dipoles by
their mutual interaction in the absence of an applied electric field. This arises from the
fact that the local field increases in proportion to the polarization. Thus, ferro-electric
materials must possess permanent dipoles. Ex.: BaTiO3, Rochelle salt
(NaKC4H4O6.4H2O), potassium dihydrogen phosphate (KH2PO4), potassium niobate
(KNbO3). These materials have extremely high dielectric constants at relatively low
applied field frequencies. Thus, capacitors made from ferro-electric materials are smaller
than capacitors made of other dielectric materials.

Piezo-electricity, or pressure electricity, is defined as polarization induced by the

application of external force. Hence, by reversing the direction of external force,
direction of the field can be reversed i.e. the application of an external electric field alters
the net dipole length and causes a dimensional change. This property is characteristic of
materials having complicated structures with a low degree of symmetry. Thus, piezo-
electric materials can be used as transducers devices that convert mechanical stress into
electrical energy and vice versa. Application for these materials includes microphones,
ultrasonic generators, sonar detectors, and mechanical strain gauges. Ex.: Barium titanate,
lead titanate, lead zirconate (PbZrO3), ammoinium dihydrogen phosphate (NH4H2PO4),
and quartz.

References

1. W. D. Kingery, H. K. Bowen, and D. R. Uhlmann, Introduction to Ceramics,

Second Edition, Wiley, New York, 1976.
2. R. E. Hummel, Electronic Properties of Materials, Third Edition, Springer-Verlag,
New York, 2000.
3. William D. Callister, Jr, Materials Science and Engineering An introduction,
sixth edition, John Wiley & Sons, Inc. 2004.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 15. Thermal properties

Engineering materials are important in everyday life because of their versatile structural
properties. Other than these properties, they do play an important role because of their
physical properties. Prime physical properties of materials include: electrical properties;
thermal properties; magnetic properties; and optical properties. Selection of materials for
use at elevated temperatures and/or temperature changes require an engineer to know and
understand their thermal properties. This chapter deals with the thermal properties of
materials. Physical property of a solid body related to application of heat energy is
defined as a thermal property.

This chapter shall describe thermal properties like heat capacity, thermal expansion,
thermal conductivity, and thermal stresses. It is important to know and understand the
concept of thermal expansion which is the root cause for thermal stresses. Thermal
stresses are stresses leading to failure of engineering structures at elevated temperatures.

15.1 Heat capacity

Many engineering solids when exposed to heat experiences an increase in temperature i.e.
it absorbs heat energy. This property of a material i.e. materials ability to absorb heat
energy is called its heat capacity, C. It is defined as the energy required to change a
materials temperature by one degree. Mathematically, it is expressed as:

dQ
C= , J/mol-K or Cal/mol-K
dT

where dQ is the energy required to produce a temperature change equal to dT.

Heat capacity is not an intrinsic property i.e. total heat a material can absorb depends on
its volume / mass. Hence another parameter called specific heat, c, it defined as heat
capacity per unit mass (J/kg-K, Cal/kg-K).
With increase of internal energy, geometrical changes may occur. Accordingly, heat
capacity is measured either at constant volume, Cv, or at constant external pressure, Cp.
The magnitude of Cp is always greater than Cv but only marginally.

Heat energy absorption of a (solid, liquid or gaseous) material exists in mode of thermal
energy vibrations of constituent atoms or molecules, apart from other mechanisms of heat
absorption such as electronic contribution. With increase of energy, atoms vibrate at
higher frequencies. However, the vibrations of adjacent atoms are coupled through
atomic bonding, which may lead to movement of lattices. This may be represented as
elastic waves (phonon) or sound waves. Vibrational contribution of heat capacity of
solids varies with temperature according to the following relation at low temperatures:

C v = AT 3

Cv is equal to zero at 0 K, but increase rapidly with temperature. It represents increased

ability of atomic vibrations or enhanced energy of lattice waves with ascending
temperature. Above a temperature called Debye temperature, d, dependence of
volumetric heat capacity value reaches saturation. This saturation values if approximately
equal to 3R ( 6 cal/mol-K), R being the universal gas constant. For many solids, value
of d is below room temperature.

Figure 15-1: Heat capacity as a function of temperature.

15.2 Thermal expansion

After heat absorption, vibrating atoms behaves as though they have larger atomic radius,
which leads to increase in materials dimensions. The phenomenon is called thermal
expansion. It is quantified in terms of thermal expansion coefficient. Linear coefficient of
thermal expansion () defined as the change in the dimensions of the material per unit
length, and is expressed as:

l f l0 l
= = =
l 0 (T f T0 ) l 0 T T

where T0 and Tf are the initial and final temperatures (in K), l0 and lf are the initial and
final dimensions of the material and is the strain. has units as (C)-1. For most metals
is the range of 5-25x10-6, for ceramics 0.5-15x10-6, and for polymers 50-400x10-6.

A volume coefficient of thermal expansion, v (=3) is used to describe the volume

change with temperature.

v
v =
v 0 T

An instrument known as dilatometer is used to measure the thermal expansion

coefficient. It is also possible to trace thermal expansion using XRD.

At microscopic level, thermal expansion can be attributed to the increase in the average
distance between the atoms. Thus, the coefficient of thermal expansion of a material is
related to the strength of the atomic bonds. The relation between inter-atomic distance
and potential energy is shown in the figure 15.2. As shown in the figure, the potential
energy curve has trough and is characterized by a minimum. This minimum corresponds
to the equilibrium inter-atomic distance ay 0 K. With increase in temperature of the
material, vibrational amplitude increase, and the mean of it represents the average inter-
atomic distance. Because of asymetric nature of the potential energy trough, rather than
the increase in atomic vibration amplitude, the average inter-atomic distance increases
with the temperature of a material.
 Figure 15.2: Change if inter-atomic distance with temperature.

If a very deep energy trough caused by strong atomic bonding is characteristic of the
material, the atoms separate to a lesser and the material has low linear coefficient of
thermal expansion. This relationship also suggests that materials having a high melting
temperature also due to strong atomic bonds have low thermal expansion coefficients.
Most ceramics having strong atomic bonds, thus, have low thermal expansion
coefficients compared with metals and polymers. However, as a consequence of non-
uniform dimensional changes, brittle materials like ceramics may experience fracture,
known as thermal shock. The capacity of material to withstand this failure is known as
thermal shock resistance. (TSR).

fk
TSR
E

where f fracture strength.

Thermal shock behavior is affected by several factors: thermal expansion coefficient a

low value is desired; thermal conductivity a high value is desired; elastic modulus
low value is desired; fracture strength high value is desired; phase transformations.
Thermal shock may be prevented by altering the external conditions to the degree that
cooling or heating rates are reduced and temperature gradients across the material are
minimized. Thermal shock is usually not a problem in most metals because metals
normally have sufficient ductility to permit deformation rather than fracture. However, it
is more of a problem in ceramics and glass materials. It is often necessary to remove
thermal stresses in ceramics to improve their mechanical strength. This is usually
accomplished by an annealing treatment.
15.3 Thermal conductivity

The ability of a material to transport heat energy from high temperature region to low
temperature region is defined as thermal conductivity. Similar to diffusion coefficient,
thermal conductivity is a microstructure sensitive property. The heat energy, Q,
transported across a plane of area A in presence of a temperature gradient T/l is given
by

T
Q = kA
l

where k is the thermal conductivity of the material. It has units as W/m.K. Metals have k
values in the range 20-400, ceramics 2-50, while polymers have in order of 0.3.

Heat energy in solids in transported by two mechanisms: lattice vibrations (phonons) and
free electrons. However, usually only one or other predominates the proceedings.
Valence electrons gain energy, move toward the colder areas of the material, and transfer
their energy to other atoms. The amount of energy transported depends on number of
excited electrons, their mobility i.e. type of material, lattice imperfections, and
temperature. The thermal energy associated with phonons is transported in the direction
of their motion.

In metals, since the valence band is not completely filled, with little thermal excitation
number of electron move and contribute to the transfer of heat energy. Thus thermal
conduction I metals is primarily due to movement of electrons. It is the same for
electrical conduction. Both conductivities are related through the following relation:

k
=L
T

where L Lorentz constant, 5.5x10-9 cal.ohm/sec.K2 or 2.44x10-8 W.ohm/ K2. The

relation is termed as Wiedemann-Franz law. Lorentz constant is supposed to be
independent of temperature and the same for all metals if the heat energy is transported
entirely by free electrons. Thus, the relationship is followed to a limited extension in
many metals.

With increase in temperature, both number of carrier electrons and contribution of lattice
vibrations increase. Thus thermal conductivity of a metal is expected to increase.
However, because of greater lattice vibrations, electron mobility decreases. The
combined effect of these factors leads to very different behavior for different metals. For
example: thermal conductivity of iron initially decreases then increases slightly; thermal
conductivity decreases with increase in temperature for aluminium; while it increases for
platinum.

In ceramics, phonons are primarily responsible for thermal conduction. Since the
electronic contribution is absent, the thermal conductivity of most ceramics is lower than
that of metals. However, main reason for experimentally observed low conductivity of
ceramics is the level of porosity, as phonons are effectively scattered by imperfections.
The scattering of phonons becomes more pronounced with rising temperature. Hence, the
thermal conductivity of ceramic materials normally diminishes with increasing
temperature. Advanced ceramic materials like AlN, SiC are good thermal conductors,
they are also electrical insulators. Therefore these materials are useful as electronic
packaging substrates where heat dissipation is needed.

Thermal conductivity of polymers is even low, compared with ceramic materials.

Vibration and movement/rotation of molecular chains transfer heat energy. In these
materials thermal conductivity depends on degree of crystallinity; a polymer with highly
crystalline and ordered structure will have higher conductivity than amorphous polymer.

15.4 Thermal stresses

Apart from thermal shock, another instance of problem exists with thermal expansion of a
material where there is no scope of dimensional changes. Thus due to temperature
changes, material may experience thermal stresses (thermal).

thermal = ET

where E elastic modulus of the material. Thermal stresses in a constrained body will be
of compressive nature if it is heated, and vice versa.

Another source for thermal stresses is thermal gradient within the body when a solid body
is heated or cooled. It is because temperature distribution will depend on its size and
shape. These thermal stresses may be established as a result of temperature gradients
across a body, which are frequently caused by rapid heating or cooling.

Engineering materials can be tailored using multi-phase constituents so that the overall
material can show a zero thermal expansion coefficient. For example: Zerodur a glass-
ceramic material that consists of 70-80% crystalline quartz, and the remaining as glassy
phase. Negative thermal expansion coefficient of glassy phase compensates for the
positive thermal expansion coefficient of the crystalline base, leading to a zero thermal
expansion. This material has many applications like parts of telescopes in astronomy.
Many ceramic materials for critical thermal applications are developed based on sodium-
zirconium-phosphate (NZP) that have a near-zero thermal expansion coefficient.

References

1. W. D. Kingery, H. K. Bowen, and D. R. Uhlmann, Introduction to Ceramics,

Second Edition, Wiley, New York, 1976.
2. William D. Callister, Jr, Materials Science and Engineering An introduction,
sixth edition, John Wiley & Sons, Inc. 2004.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 16. Magnetic properties

Engineering materials are important in everyday life because of their versatile structural
properties. Other than these properties, they do play an important role because of their
physical properties. Prime physical properties of materials include: electrical properties;
thermal properties; magnetic properties; and optical properties. Magnetic properties play
an important role in daily life. Magnetic materials are used in electric motors, generators,
transformers. Modern-day devices use data storage technology that is based on magnetic
particles. Magnetic materials are used in devices like telephones, televisions,
supercomputers, etc. they are also used in medical technology, for example DNA
sequencing.

This chapter shall describe magnetic properties like dia-, para-, and ferro-magnetism
along with anti-ferro- and ferri- magnetism. It also includes discussion about temperature
effects on magnetic behavior, followed by magnetic domains and hysteresis.

Magnetism is a phenomenon by which a material exerts either attractive or repulsive

force on another. Basic source of magnetic force is movement of electrically charged
particles. Magnetic behavior of a material can be traced to the structure of atoms.
Electrons in atoms have a planetary motion in that they go around the nucleus. This
orbital motion and its own spin cause separate magnetic moments, which contribute to
the magnetic behavior of materials. Thus every material can respond to a magnetic field.
However, the manner in which a material responds depend much on its atomic structure,
and determines whether a material will be strongly or weakly magnetic.

Magnetic moment due to spin of an electron is known as Bohr magneton,MB, is the most
B

fundamental magnetic moment.

qh
MB = = 9.274 X 10 24 A.m 2
4me

where q is the charge on the electron, h Plancks constant, me mass of electron. This
moment is directed along the axis of electron spin. Magnetic moment resulted from
particles in nucleus is much smaller than that result from spin of electron, so it is ignored.
If magnetic moment of electrons could sum up, then the world would be a magnetic
place. Fortunately, two reasons are found to explain this phenomenon:- First: according
to Pauli exclusion rule, two electrons with same energy level must have opposite spins
thus so are their magnetic moments, which cancel out each other. Second: orbital
moments of electrons also cancel out each other thus no net magnetic moments if there
is no unpaired electron(s). Some elements such as transition elements, lanthanides, and
actinides have a net magnetic moment since some of their energy levels have an unpaired
electron.

Magnetic dipoles are found to exist in magnetic materials, analogous to electric dipoles.
A magnetic dipole is a small magnet composed of north and south poles instead of
positive and negative charges. Within a magnetic field, the force of field exerts a torque
that tends to orient the dipoles with the filed. Magnetic forces are generated by moving
electrically charged particles. These forces are in addition to any electrostatic forces that
may already exist. It is convenient to think magnetic forces in terms of distributed field,
which is represented by imaginary lines. These lines also indicate the direction of the
force. If a magnetic field is generated by passing current I through a coil of length l and
number of turns n, then the magnetic field strength is given by:

nI
H =
l

Units for magnetic field strength, thus, are A/m.

Magnetic flux density (induction) is the measure of lines within a medium. It has units as
weber (Wb) /m2 or tesla.

B = H

where permeability. It is a specific property of the medium, and has units as Wb/A.m
or henries (H) /m.

Flux density, B, is determined by the manner in which induced and permanent dipoles
interact with the applied field. If the magnetic moments reinforce the applied field i.e.
greater number of lines of flux are created, and the field is magnified. This is represented
as

> 0

where 0 magnetic permeability of vacuum.

Several parameters may be used to describe magnetic properties of solids. One of them is
relative permeability (r). It is a measure of the degree to which the material can be
magnetized.

r =
0

Another field quantity called magnetization, M, is defined as

B = 0 H + 0 M = 0 r H

M = mH

m is called the magnetic susceptibility.

m = r 1

There are many ways a material can be magnetized i.e. many types of magnetism. Three
basic magnetisms are: dia-magnetism, para-magnetism and ferro-magnetism. Anti-ferro-
magnetism and ferri-magnetisms are considered as subclasses of ferro-magnetism. A
material exhibits one of these magnetisms.

16.1 Dia-, Para-, and Ferro-magnetism

Dia-magnetism is very weak form of magnetism. It exists only when an external field is
applied, and is non-permanent. The applied external field acts on atoms of a material,
slightly unbalancing their orbiting electrons, and creates small magnetic dipoles within
atoms which oppose the applied field. This action produces a negative magnetic effect
known as diamagnetism. The induced magnetic moment is small, and the magnetization
(M) direction is opposite to the direction of applied field (H). Thus the relative
permeability is less than unity i.e. magnetic susceptibility is negative, and is in order of -
10-5. Dia-magnetism is virtually found in all materials; however it is observable only in
absence of other magnetisms. This form of magnetism is of no practical importance.
Materials such as Cu, Ag, Si, Ag and alumina are diamagnetic at room temperature.
Superconductors are perfect dia-magnets (m=-1); they lose their superconductivity at
higher temperatures or in the presence of a magnetic field.

Materials which exhibit a small positive magnetic susceptibility in the presence of a

magnetic field are called para-magnetic, and the effect is termed as para-magnetism.
When materials have unpaired electrons, a net magnetic moment due to electron spin is
associated with each atom. In the absence of an external field, the orientations of these
atomic magnetic moments are random leading to no net magnetization. When an external
field is applied dipoles line-up with the field, resulting in a positive magnetization.
However, because the dipoles do not interact, extremely large magnetic fields are
required to align all of the dipoles. In addition, the effect is lost as soon as the magnetic
field is removed. Since thermal agitation randomizes the directions of the magnetic
dipoles, an increase in temperature decreases the paramagnetic effect. Para-magnetism is
produced in many materials like aluminium, calcium, titanium, alloys of copper.
Magnetic susceptibility of these materials is slightly positive, and lies in the range 10-5 to
10-2.

Both dia- and para- magnetic materials are considered as non-magnetic because they
exhibit magnetization only in presence of an external field. Certain materials possess
permanent magnetic moments even in the absence of an external field. This is result of
permanent unpaired dipoles formed from unfilled energy levels. These dipoles can easily
line-up with the imposed magnetic field due to the exchange interaction or mutual
reinforcement of the dipoles. These are chrematistics of ferro-magnetism. Materials with
ferro-magnetism (Examples: Fe, Co, Ni, Gd) possess magnetic susceptibilities
approaching 106. Consequently,

H << M , and B 0 M

The mutual spin alignment in ferro-magnetic materials exists over relatively large volume
regions of the crystal called domains. The susceptibility of ferro-magnetic materials
depends upon the intensity of the applied field. Above the Curie temperature, ferro-
magnetic materials behave as para-magnetic materials and their susceptibility is given by
the Curie-Weiss law, defined as follows:

C
m =
T Tc

where C material constant, T temperature, Tc Curie temperature.

In some materials such as Mn, Cr, MnO, NiO, CoO, MnCl2 the magnetic moments
produced in neighboring dipoles line up in opposition to one another in the magnetic
field, even though the strength of each dipole is very high. This will result in zero
magnetization, and the effect is called anti-ferro-magnetism. Exchange interaction which
is responsible for parallel alignment of spins is extremely sensitive to inter-atomic
spacing and to the atomic positions. This sensitivity causes anti-parallel alignment of
spins. When the strength of anti-parallel spin magnetic moments is equal, no net spin
moment exists, and resulting susceptibilities are quite small. One noticeable characteristic
of anti-ferro-magnets is they attain maximum susceptibility at a critical temperature
called Neel temperature. At temperatures above this, anti-ferro-magnets become para-
magnetic.

On the other hand, some ceramic materials exhibit net magnetization. This is because of
either the strength or number of opposing dipoles is not equal. In a magnetic field, the
dipoles of a cation may line up with the field, while dipoles of other cation may not.
These ceramics are called ferrites, and the effect is known as ferri-magnetism. Ferri-
magnetism is similar to anti-ferro-magnetism in that the spins of different atoms or ions
line up anti-parallel. However, the spins do not cancel each other out, and a net spin
moment exists. Below the Neel temperature, therefore, ferromagnetic materials behave
very much like ferromagnetic materials and are paramagnetic above the Neel
temperature. These materials exhibit a large but field dependent magnetic susceptibility
similar to ferro-magnets. They also show Curie-Weiss behavior. As these ceramics are
good insulators, electrical losses are minimal, and hence ferrites have lot of applications
in devices such as high frequency transformers. Examples: Fe3O4, NiFe2O4,
(Mn.Mg)Fe2O4, PbFe12O19, Ba Fe12O19, YIG yttrium iron garnet Y3Fe5O12.

The table 16.1 compares different magnetism.

Table 16-1: Various types of magnetisms.

Magnetic susceptibility
Magnetism Examples
sign magnitude
Organic materials, superconducting
Dia - Small, Constant
materials, metals like Bi
Alkali and transition metals, rare
Para + Small, Constant
earth elements
Transition metals (Fe, Ni, Co), rare
Ferro + Large, Function of H
earth elements (Gd)
Anti-Ferro + Small, Constant Salts of transition elements (MnO)
Ferrites (MnFe2O4, ZnFe2O4) and
Ferri + Large, Function of H
chromites

16.2 Influence of temperature on magnetic behavior

Temperature does influence the magnetic characteristics of materials, as it influences

electrical properties. With rising temperature, magnitude of the atom thermal vibrations
increases. This may lead to more randomization of atomic magnetic moments as they are
free to rotate. Usually, atomic thermal vibrations counteract forces between the adjacent
atomic dipole moments, resulting in dipole misalignment up to some extent both in
presence and absence of external field. As a consequence of it, saturation magnetization
initially decreases gradually, then suddenly drops to zero at a temperature called Curie
temperature, Tc. The magnitude of the Curie temperatue is dependent on the material. For
example: for cobalt 1120 C, for nickel 335 C, for iron 768 C, and for Fe3O4 585
C.

16.3 Domains and Hysteresis

In addition to susceptibility differences, the different types of magnetism can be

distinguished by the structure of the magnetic dipoles in regions called domains. Each
domain consists of magnetic moments that are aligned, giving rise to a permanent net
magnetic moment per domain. Each of these domains is separated from the rest by
domain boundaries / domain walls. Boundaries, also called Bolch walls, are narrow zones
in which the direction of the magnetic moment gradually and continuously changes from
that of one domain to that of the next. The domains are typically very small about 50 m
or less, while the Bloch walls are about 100 nm thick. For a polycrystalline specimen,
each grain may have more than one microscopic sized domain.

Domains exist even in the absence of an external field. In a material that has never been
exposed to a magnetic field, the individual domains have a random orientation. This type
of arrangement represents the lowest free energy. The domain structure of material is
determined by many types of energies, with most stable structure being attained when the
overall potential energy of the material is a minimum. Total magnetic energy of a
material is the sum of the contributions of the following: exchange energy, magneto-
static energy, magneto-crystalline anisotropy energy, domain wall energy and magneto-
strictive energy. When the bulk material is un-magnetized, the net magnetization of these
domains is zero, because adjacent domains may be orientated randomly in any number of
directions, effectively canceling each other out.

The average magnetic induction of a ferro-magnetic material is intimately related to the

domain structure. When a magnetic field is imposed on the material, domains that are
nearly lined up with the field grow at the expense of unaligned domains. This process
continues until only the most favorably oriented domains remain. In order for the
domains to grow, the Bloch walls must move, the external field provides the force
required for this moment. When the domain growth is completed, a further increase in the
magnetic field causes the domains to rotate and align parallel to the applied field. At this
instant material reaches saturation magnetization and no further increase will take place
on increasing the strength of the external field. Saturation magnetization is the greatest
amount of magnetization tat the material can obtain. Under these conditions the
permeability of these materials becomes quite small. Figure 16-1 in the following
presents the relation between the applied field strength and magnetization of the material.

Figure 16-1: Magnetization saturation of a material with applied field.

As shown in figure 16-1, magnetization increases with applied filed, and reaches a
saturation value in a sufficiently stronger filed. Once magnetic saturation has been
achieved, a decrease in the applied field back to zero results in a macroscopically
permanent or residual magnetization, known as remanance, Mr. The corresponding
induction, Br, is called retentivity or remanent induction of the magnetic material. This
B

effect of retardation by material is called hysteresis. The material acts as a permanent

magnet, even at zero applied field. At this point, spin orientations within domains have
readily rotated back to their favorable position, but the original random domain
arrangement is not achieved. This is because the resistance offered by the domain walls
prevents re-growth of the domains i.e. domain growth process is not entirely reversible,
and domain wall motion is limited. The magnetic field strength needed to bring the
induced magnetization to zero is termed as coercivity, Hc. This must be applied anti-
parallel to the original field.

A further increase in the field in the opposite direction results in a maximum induction
i.e. saturation magnetization, but in the opposite direction. The field can once again be
reversed, and the field-magnetization loop can be closed, and so the corresponding field-
induction loop. This loop is known as hysteresis loop or B-H plot or M- H plot. The area
within the hysteresis loop represents the energy loss per unit volume of material for one
cycle. The B-vs-H curve in figure 16-2 represents a hysteresis loop taken to saturation.
However, it is not necessary to increase the field strength to saturation to generate the
loop.

Figure 16-2: Schematic presentation of a typical hysteresis loop.

The coercivity of the material is a micro-structure sensitive property. This dependence is

known as magnetic shape anisotropy. The coercivity of recording materials needs to be
smaller than that for others since data written onto a data storage medium should be
erasable. On the other hand, the coercivity values should be higher since the data need to
be retained. Thus such materials are called magnetically semi-hard. Examples: Hard
ferrites based on Ba, CrO2, -Fe2O3; alloys based on Co-Pt-Ta-Cr, Fe-Pt and Fe-Pd, etc.
Soft magnets are characterized by low coercive forces and high magnetic permeabilities;
and are easily magnetized and de-magnetized. They generally exhibit small hysteresis
losses. Application of soft magnets include: cores for electro-magnets, electric motors,
transformers, generators, and other electrical equipment. Examples: ingot iron, low-
carbon steel, Silicon iron, superalloy (80% Ni-5% Mo-Fe), 45 Permalloy (55%Fe-
45%Ni), 2-79 Permalloy (79% Ni-4% Mo-Fe), MnZn ferrite / Ferroxcube A (48%
MnFe2O4-52%ZnFe2O4), NiZn ferrite / Ferroxcube B (36% NiFe2O4-64% ZnFe2O4), etc.

Hard magnets are characterized by high remanent inductions and high coercivities. These
are also called permanent magnets or hard magnets. They generally exhibit large
hysteresis losses. Examples: Co-steel, Tungsten steel, SmCo5, Nd2Fe14B, ferrite
Bao.6Fe2O3, Cunife (60% Cu 20% Ni-20% Fe), Alnico (alloy of Al, Ni, Co and Fe), etc.
Applications include fractional horse-power motors, automobiles, audio- and video-
recorders, earphones, computer peripherals, and clocks.

References

1. William D. Callister, Jr, Materials Science and Engineering An introduction,

sixth edition, John Wiley & Sons, Inc. 2004.
2. V. Raghavan, Materials Science and Engineering, third edition, Prentice Hall of
India Private Limited, New Delhi, 1990.
3. D. Jiles, Introduction to Magnetism and Magnetic Materials, Nelson Thornes,
Cheltenham, UK 1998.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 17. Optical properties

Engineering materials are important in everyday life because of their versatile structural
properties. Other than these properties, they do play an important role because of their
physical properties. Prime physical properties of materials include: electrical properties;
thermal properties; magnetic properties; and optical properties. Optical properties play an
important role in daily life. They have had a significant impact on the development of the
communications infrastructure and the information technology. They are also useful in
fields like medicine, manufacturing, astronomy, etc.

The goal of this chapter is to present basic concepts about optical properties, optical
properties of metals and non-metals, and finally applications of optical phenomena.

17.1 Basic concepts

Optical property of a material is related to the interaction of it with electromagnetic

radiation. This radiation may have characteristics that fall in the visible light spectrum, or
may be even out of it. Electromagnetic spectrum of radiation spans the wide range from
-rays with wavelength as 10-12 m, through x-rays, ultraviolet, visible, infrared, and
finally radio waves with wavelengths as along as 105 m. Visible light is one form of
electromagnetic radiation with wavelengths ranging from 0.39 to 0.77 m. It contains
color bands from violet through red, as shown in the figure 17.1. White light is simply a
mixture of all colors. The ultraviolet region covers the range from about 0.01 to about
0.40 m, and the infrared region extends from about 0.75 to 1000 m.
 Figure-17.1: Wave length spectrum of electro-magnetic waves.

The true nature of the light will probably never be known. However, light can be
considered as having waves and consisting of particles called photons. The important
characteristics of the photons are related by the following equation. This equation allows
us to consider the photon as a particle of energy, or as a wave with a characteristic
wavelength and frequency.

hc0
E = h =

where E energy, h Plancks constant (6.62x10-34 J.sec), frequency, c0 speed of

light in vacuum (3x108 m/sec), and wavelength.

All materials interact in some way with light. Interaction of photons with the electronic or
crystal structure of a material leads to a number of phenomena. The photons may give
their energy to the material (absorption); photons give their energy, but photons of
identical energy are immediately emitted by the material (reflection); photons may not
interact with the material structure (transmission); or during transmission photons are
changes in velocity (refraction).

At any instance of light interaction with a material, the total intensity of the incident light
striking a surface is equal to sum of the absorbed, reflected, and transmitted intensities
i.e.

I 0 = I A + I R + IT
The intensity is defined as the number of photons impinging on a surface per unit area per
unit time. Materials that are capable of transmitting light with relatively little absorption
and reflection are called transparent materials i.e. we can see through them. Translucent
materials are those through which light is transmitted diffusely i.e. objects are not clearly
distinguishable when viewed through. Those materials that are impervious to the
transmission of visible light are termed as opaque materials.

17.2 Optical properties of metals and alloys

Typical characteristic of metals with respect to crystal structure is that they possess a
high-energy band that is only partially filled with electrons. When visible light in directed
on a metal surface, the energy is used to excite electrons into unoccupied energy states
above the Fermi level, thus making metals behave as opaque materials i.e. light is
absorbed. Except of thin sections, metals strongly reflect and/or absorb incident radiation
for long wavelengths to the middle of the ultraviolet range i.e. metals are opaque to all
electromagnetic radiation on the low end of the frequency spectrum, from radio waves,
through infrared, visible, into middle of the ultraviolet radiation. However, metals are
transparent to high end frequencies, ex. x-ray and ray radiation. Total absorption by
metals is within a very thin outer layer, usually less than 0.1 m; thus only metallic films
thinner than 0.1 m are capable of transmitting visible light.

Most of the absorbed radiation is emitted from the metallic surface in the form of visible
light of the same wavelength as reflected light. The reflectivity of metals is about 0.95,
while the rest of impinged energy is dissipated as heat. The amount of energy absorbed
by metals depends on the electronic structure of each particular metal. For example: with
copper and gold there is greater absorption of the short wavelength colors such as green
and blue and a greater reflection of yellow, orange and red wavelengths. Other metals
such as silver and aluminium strongly reflect all parts of the visible spectrum and show a
white silvery color.

17.3 Optical properties of non-metallic materials

By virtue of their electron structure with characteristic energy band structures, non-
metallic materials may be transparent to visible lights. Thus, all four optical phenomena
such as absorption, reflection, transmission and refraction are important for these
materials.

Refraction: When light photons are transmitted through a material, they causes
polarization of the electrons in the material and by interacting with the polarized
materials, photons lose some of their energy. As a result of this, the speed of light is
reduced and the beam of light changes direction.

The relative velocity of light passing through a medium is expressed by the optical
property called the index of refraction (n), and is defined as:
 c0
n=
c

where c0 speed of light in vacuum, c speed of light in the concerned material. If the
angle of incidence from a normal to the surface is i, and the angle of refraction is r, the
refractive index of the medium, n, is given by

sin i
n=
sin r

provided that the incident light is coming from a phase of low refractive index such as
vacuum or air. Reflection and refraction of light is shown in figure 17.2.

Figure-17.2: Reflection and refraction of light.

The speed of light in a material can be related to its electrical and magnetic properties as:

1
c=
.

where electrical permittivity, and magnetic permeability. Thus,

c0 .
n= = = r . r
c 0 . 0

Since most substances are only slightly magnetic i.e. r 1, then

n r

Thus, for transparent materials, index of refraction and dielectric constant are related.
Refractive indices of some materials are listed in the table 17.1.

Table-17.1: Index of refraction for typical materials.

 Material Refractive index Material Refractive index
Air 1.00 Epoxy 1.58
Ice 1.309 Polystyrene 1.60
Water 1.33 Spinel, MgAl2O3 1.72
Teflon 1.35 Sapphire, Al2O3 1.76
Silica glass 1.458 Rutile, TiO2 2.68
Polymethyl 1.49 Diamond 2.417
methacrylate
Silicate glass 1.50 Silicon 3.29
Polyethylene 1.52 Gallium arsenide 3.35
NaCl 1.54 Germanium 4.00

Snells law of light refraction refractive indices for light passing through from one
medium with refractive index n through another of refractive index n is related to the
incident angle, , and refractive angle, , by the following relation

n sin '
=
n' sin

If light passes from a medium with a high refractive index to one with a low refractive
index, there is a critical angle of incidence, c, which if increased will result in total
internal reflection of the light. This angle is defined as (refraction) = 90.

Reflection: When a beam of photons strikes a material, some of the light is scattered at
the interface between the two media even if both are transparent. Reflectivity, R, is a
measure of fraction of incident light which is reflected at the interface, and is given by

IR
R=
I0

Where I0 and IR are the incident and reflected bean intensities respectively. If the material
is in a vacuum or in air:

2
n 1
R=
n +1

If the material is in some other medium with an index of refraction of ni, then:

2
n ni
R =
n + ni

The above equations apply to the reflection from a single surface and assume normal
incidence. The value of R depends upon the angle of incidence. Materials with a high
index of refraction have a higher reflectivity than materials with a low index. Because the
index of refraction varies with the wavelength of the photons, so does the reflectivity. In
metals, the reflectivity is typically on the order of 0.90-0.95, whereas for glasses it is
close to 0.05. The high reflectivity of metals is one reason that they are opaque. High
reflectivity is desired in many applications including mirrors, coatings on glasses, etc.

Absorption: When a light beam in impinged on a material surface, portion of the incident
beam that is not reflected by the material is either absorbed or transmitted through the
material. The fraction of beam that is absorbed is related to the thickness of the materials
and the manner in which the photons interact with the materials structure. Thus,
according to Bouguers law:

I = I 0 exp( .x)

where I intensity of the beam coming out of the material, I0 intensity of the incident
beam, x path through which the photons move, and linear absorption coefficient,
which is characteristic of a particular material.

Absorption in materials occurs mainly by two mechanisms (1) Rayleigh scattering

where photon interacts with the electrons orbiting an atom and is deflected without any
change in photon energy. This is significant for high atomic number atoms and low
photon energies. Ex.: Blue color in the sunlight gets scattered more than other colors in
the visible spectrum and thus making sky look blue. Tyndall effect is where scattering
occurs from particles much larger than the wavelength of light. Ex.: Clouds look white.
(2) Compton scattering here incident photon knocks out an electron from the atom
loosing some of its energy during the process. This is also significant for high atomic
number atoms and low photon energies. Photoelectric effect occurs when photon energy
is consumed to release an electron from atom nucleus. This effect arises from the fact that
the potential energy barrier for electrons is finite at the surface of the metal. Absorption
occurs at particular levels of photon energies, which are equal to that of binding energies.
The energy at which this occurs is called the absorption edge.

Transmission: the fraction of beam that is not reflected or absorbed is transmitted through
the material. Thus the fraction of light that is transmitted through a transparent material
depends on the losses incurred by absorption and reflection. Thus,

R + A+T =1

where R reflectivity, A aborptivity, and T transmitivity. Each of these parameters

are characteristic of material, and they also depend on light wavelength.

If the incident light is of intensity I0, then the loss due to reflection at the front end of the
material is RI0. Thus the fraction of beam intensity entering the material is

I after .reflection = (1 R ) I 0
Once the beam enters the material, a portion of it is absorbed. Thus

I after .absorption = (1 R ) I 0 exp( .x)

Before the beam exits at the back surface, a portion of it will be reflected again. Thus

I after .reflection .at .back .surface = R (1 R ) I 0 exp( .x )

Thus, the fraction of beam that is actually transmitted through the material is given by

I transmitte d = I after .absorption I after .reflection .at .back .surface = (1 R ) I 0 exp( .x ) R (1 R ) I 0 exp( .x )
Thus, I t = I 0 (1 R ) 2 exp( .x )

The process of light transmission is shown schematically in the figure 17.3.

Figure-17.3: Light transmission.

17.4 Optical phenomena

As mentioned in earlier sections, light interacts with a material in many ways. Depending
on the material, its crystal-/micro-structure, and also on the characteristics of incident
light, there are many peculiar phenomena occurs, which are known as optical
phenomena. These include: luminescence, lasers, thermal emission, photo-conductivity,
and optical fibers. All these find quite many applications in technology for every day life.

Luminescence: Luminescence is defined as the process in which a material absorbs

energy and then immediately emits visible or near-visible radiation. This occurs as a
result of excitation of electrons of a material from the valence band into the conduction
band. The source of input energy may be high energy electrons or light photons. During
luminescence, the excited electrons drop back to lower energy levels. If the emission
takes place within 10-8 sec.s after excitation, the luminescence is called fluorescence, and
if it takes longer than 10-8 sec.s, it is known as phosphorescence.

Luminescence takes place in outer valence- and conduction- bands, while x-rays are
produced during electron transitions in the inner-energy levels of an atom. Luminescence
does not occur in metals. In certain ceramics and semi-conductors, however, the energy
gap between the valence and conduction bands is such that an electron dropping through
this gap produces a photon in the visible range.

Ordinarily pure materials do not display this phenomenon. Special materials called
phosphors have the capability of absorbing high-energy radiation and spontaneously
emitting lower-energy radiation. Ex.: some sulfides, oxides, tungstates, and few organic
materials. The emission spectra of these are controlled by added impurities referred as
activators which provide discrete energy levels in the gap between valence and
conduction bands. The excited electrons first drop to donor level, and get trapped. When
the source is removed, they gradually escape the trap and emit light over some additional
period of time. The intensity of the luminescence is given by

t
I = I 0 exp( )

where I0 initial intensity of luminescence, I fraction of luminescence after time, t, -

relaxation time, constant for a material.

Luminescence process is classified based on the energy source for electron excitation as
photo-luminescence, cathode-luminescence, and electro-luminescence.

Photo-luminescence occurs in fluorescent lamps. Here ultra-violet radiation from low-

pressure mercury arc is converted to visible light by calcium halo-phosphate phosphor
(Ca10F2P6O24). In commercial lamps, about 20% of F- ions are replaced with Cl- ions.
Antimony, Sb3+, ions provide a blue emission while manganese, Mn2+, ions provide an
orange-red emission band.

Cathode-luminescence is produced by an energized cathode which generates a beam of

high-energy bombarding electrons. Applications of these include: electron microscope;
cathode-ray oscilloscope; color television screens. Here, however, relaxation time must
not be too long which makes the picture blur or they may even overlap. The modern
televisions have very narrow, about 0.25 mm wide, vertical stripes of red-, green-, and
blue- emitting phosphors deposited on the inner surface of the screens. With help of steel
shadow mask with small elongated holes, the incoming signal is scanned over the entire
screen at 30 times per second. The small and large number of phosphors consecutively
exposed in the rapid scan of 15750 horizontal lines per second. The persistence of human
eye makes possible a clear visible picture with good resolution. Commercial phosphors
for different colors are: red yttrium oxy-sulfide (Y2O2S) with 3% europium (Eu); green
(Zn,Cd)S with a Cu+ acceptor and Al3+ donor; blue zinc sulfide (ZnS) with Ag+
acceptor and Cl- donor.

Electro-luminescence occurs in devices with p-n rectifying junctions which are

stimulated by an externally applied voltage. When a forward biased voltage is applied
across the device, electrons and holes recombine at the junction and emit photons in the
visible range (mono-chromatic light i.e. singe color). These diodes are called light
emitting diodes (LEDs). The characteristic color of an LED depends on the particular
semi-conducting material that is used. GaAs, GaP, GaAlAs, and GaAsP are typical
materials for LEDs. LEDs emit light of many colors, from red to violet, depending on the
composition of the semiconductor material used. Some even emit light outside of the
visible spectrum, i.e., infrared and ultraviolet. The following table 17.2 lists materials
used in different colored LEDs.

Table-17.2: Materials for colored LEDs.

Wave length (nm) Color Material

- Infra-red GaAs
660 Red GaP0.40As0.60 or Al0.25Ga0.75As
635 Orange GaP0.65As0.35
578 Yellow GaP0.85As0.15
556 Green GaP (GaP1.00As0.00)
- Blue Ga0.94NIn0.06

Lasers: Laser is an acronym for light amplification by stimulated emission of radiation. It

is in fact special application of luminescence. Unlike most radiation processes, such as
luminescence, which produce incoherent light, the light produced by laser emission is
coherent i.e. light waves are all in phase with each other. Consequently, laser light waves
are does not spread out i.e. parallel, directional, and monochromatic i.e. entirely of one
wavelength.

In certain materials, electrons excited by a stimulus produce photons which in turn excite
additional photons of identical wavelength. Thus a large amplification of the photons
emitted in the material occurs. By selecting stimulant and material properly, laser beam
can be in the visible range. Lasers are useful in many applications such as welding, metal
cutting, heat treatment, surgery, mapping, reading compact disks, etc. A variety of
materials are used to produce lasers. Ex.: Ruby, single crystal of Al2O3 doped with little
amount of Cr2O3; yttrium aluminium garnet (Y3Al5O12 YAG) doped with neodymium,
Nd; CO2 gas; He-Ne gas; some semi-conductors like GaAs and InGaAsP. Gas lasers
generally produce lower intensities and powers, but are more suitable for continuous
operation since solid-state lasers generate appreciable amounts of heat.

Laser operation: when the laser material is exposed to stimulant, for example flash lamp,
electrons that initially fills the lowest-energy levels gets excited into higher energy-levels.
These electrons can decay back by two paths: one in which they fall directly back
associated photon emissions are not part of the laser beam; others decay into a
intermediate meta-stable state where they reside for about 3 ms before spontaneous
emission. This initial spontaneous emission acts as stimulus and triggers an avalanche of
emissions from remaining electrons in the meta-stable state. The photons are of the same
energy and are in phase. The beam is collimated through the use of a tube with silvered
mirrors at each end. As stimulated emission occurs, only those photons traveling nearly
parallel to the log axis of the material are reflected. These reflected photons stimulate the
emission of more photons. Reflected photons traveling up and down the length of the
crystal produce ever-increasing number of stimulated photons. Finally, high-energy,
highly-collimated, monochromatic beam of coherent light is emitted from the laser
device.

Thermal emission: When a material is heated, electrons are excited to higher energy
levels, particularly in the outer energy levels where the electrons are less strongly bound
to the nucleus. These excited electrons, upon dropping back to the ground state, release
photons in process what is called thermal emission.

During thermal emission a continuous spectrum of radiation is emitted with a minimum

wavelength and the intensity distribution is dependent on the temperature. Higher the
temperature, wider will be the range of wavelengths emitted. By measuring the intensity
of a narrow band of the emitted wavelengths with a pyrometer, materials temperature
can be estimated.

Photo-conductivity: As mentioned in earlier section upon absorption of photons at

surface, electron may be released from its atom nucleus. Thus electricity can be generated
from the surface of a metal when it is bombarded with photons. Similarly, bombardment
of semiconductors by photons, with energy equal to greater than the band gap, may result
in creation of electron-hole pairs that can be used to generate current. This process is
called photo-conductivity, and is different from photo-electric effect in the sense that an
electron-hole pair is generated whose energy is related to the band gap energy instead of
free electron alone whose energy is related to the Fermi level. The current produced in
photo-conductivity is directly related to the incident light intensity.

This phenomenon is utilized in photographic light meters. Cadmium sulfide (CdS) is

commonly used for the detection of visible light, as in light meters. Photo-conductivity is
also the underlying principle of the photo-voltaic cell, known to common man as solar
cell, used for conversion of solar energy into electricity.

Optical fibers: Recently the buzz word in the communications sector is the optical fiber,
using in place of metallic copper wires. Signal transmission through a metallic wire
conductor is electronic, whereas in fibers it is photonic i.e. by photons. This enables
faster transmission at higher densities to longer distances with reduction in error rate.
These systems consists of transmitter (a semiconductor laser) to convert electrical signals
to light signals, optical fiber to transmit the light signals, and a photodiode to convert
light signals back to electrical signals.
Optical fiber is the heart of the communication system. It must have extremely low loss
of light, must be able to guide the light pulses over long distances without significant loss
and/or distortion. It primarily consists of core, cladding and coating. The core transmits
the signals, while the cladding constrains the light beam to the core; outer coating
protects the core and cladding from the external environment. Optical fiber operates on
the principle of total internal reflectance. Typically both the core and cladding are made
of special types of glass with carefully controlled indices of refraction. The indices of
refraction are selected such that

ncladding < ncore

Once the light enters the core from the source, it is reflected internally and propagates
along the length of the fiber. Internal reflection is accomplished by varying the index of
refraction of the core and cladding glass materials. Usually two designs are employed in
this regard. In step-index optical fiber, there is a sharp change in refractive index between
the core and cladding. In this design output pulse will be broader than the input one. It is
because light rays traveling in different trajectories have a variety of path lengths. It is
possible to avoid pulse broadening by using graded-index fiber. This results in a helical
path for the light rays, as opposed to zig-zag path in a step-index fiber. Here impurities
such as boron oxide (B2O3) or germanium dioxide /GeO2) are added to the silica glass
such that the index of refraction varied gradually in parabolic manner across the cross
section. This enables light to travel faster while close to the periphery than at the center.
This avoids pulse broadening i.e. light rays arrive at output at approximately same time.
Both step- and graded- index fibers are termed as multi-mode fibers. Third type optical
fiber is called single-mode fiber in which light travels largely parallel to the fiber axis
with little distortion of the digital light pulse. These are used for long transmission lines.

Core and cladding materials are selected not only on the basis of their refractive indices,
but also on basis of ease of manufacturability, light loss, mechanical strength properties
and dispersion properties. However, density () and refractive index (n) are critical.
These two parameters are related approximately as

+ 10.4
n=
8.6

High-purity silica-based glasses are used as fiber material, with fiber diameter ranging
from 5 to 100 m. The fibers are carefully fabricated to be virtually free from flaws, as a
result, are extremely strong and flexible. Uniformity of fiber cross-sectional dimensions
and core roundness are critical; allowable tolerances of these parameters are within a
micro-meter over 1 km of length.

References

1. William D. Callister, Jr, Materials Science and Engineering An introduction,

sixth edition, John Wiley & Sons, Inc. 2004.
2. K. M. Ralls, T. H. Courtney, and J. Wulff, Introudction to Materials Science and
Engineering, Wiley, New York, 1976.
1. W. D. Kingery, H. K. Bowen, and D. R. Uhlmann, Introduction to Ceramics,
Second Edition, Wiley, New York, 1976.
 Material Science

Prof. Satish V. Kailas

Associate Professor
Dept. of Mechanical Engineering,
Indian Institute of Science,
Bangalore 560012
India

Chapter 18. Economic, Environmental and Social issues of

material usage

Engineering materials are important in everyday life because of their versatile structural
and physical properties. Selection of the engineering materials is dependent on these
properties. However, choice of material has implications throughout the life-cycle of a
product, influencing many aspects of economic and environmental performance. In other
words, different material issues are needed to be dealt with during material selection if
the product is to be commercially successful and competitive in the market. These special
considerations are related to their cost, production cost, environmental and social issues
like pollution, disposal, recycling, etc. This chapter deals with these special issues of
material selection.

18.1 Economic considerations

Engineering profession deals with utilization of scientific and technological advances to

design and manufacture components and systems that perform reliably and satisfactorily.
However, there are economics as the driving force behind it. Economics of engineering a
component / system depends on three factors: component design, material usage, and
manufacturing costs. These three factors are inter-related in the sense that one or two
might influence the choice of others.

Initially any component / system need to go through conception and then design stage.
This includes generation of concept about the component / system. Later on design stage
takes care of its size, shape, and configuration which will influence the performance of it
during the service. Usually engineers deal with not a single component, but with complex
assembly of components / a system. Thus, each component needs to be designed for
greater efficiency of the system. This may some times act as constraint to optimal design
of a component. Hence, design of a component is usually an iterative process. Less the
number of iterations, lower will be the cost of the component / system.

Design stage is followed by material selection. Material is selected depending on its

properties, which are suitable to serve the purpose during the service. Other than the
properties, cheaper materials are preferred, if choice is available. Thus, usually a family
of materials is selected that satisfy the design constraints, then comparisons are made on
the basis of cost per unit. This cost also includes unavoidable material wastage during
manufacturing stage.

After design and material selection, it is up to manufacturing method to reduce the

product cost. Usually manufacturing includes both primary stage and secondary stage.
The cost considerations include the capital on tooling, maintenance cost, labor, repair
costs, and material wastage. More the number of manufacturing stages, higher will be the
product cost. Inspection, assembly, and final packaging will add-on to the product cost.

18.2 Environmental and Social considerations

Manufacturing of a product does have impact on environment and also on society in

many ways. This is because resources required to produce a product comes from different
parts of the world, and the industrial wealth and prosperity are shared by more the one
region. Along with these, detrimental effects of industrialization also spread its wings to
various parts of the world.

A material used to produce a product goes through number of stages / phases. These
include extraction of raw materials from natural resources through production, use during
the service, and finally its disposal. It is some times known as cradle-to-grave life cycle
of a material. A schematic life cycle model has been presented in the figure 18.1.

Figure-18.1: Life cycle model.

As depicted in the above figure, raw materials are first extracted from natural earthy
resources through drilling, mining, etc. later-on these are subjected to purification,
refining to convert them into metals, ceramics, rubber, fuel, etc. These primary products
are further processed to obtain engineered materials like metallic alloy, glass, plastics,
semi-conductors, etc. Now the engineered materials are shaped, heat treated to make
components which are assembled into products, devices that are ready for use by society.
During the service, products become old, out fashioned, break down, or may not serve the
purpose efficiently. So they are discarded. Upon this, they go as waste into land fills, or
some times recycled and some components may be reused. This competes the life cycle.

Earthy resources are, unfortunately, limited. However, some are renewable while
majority are not! As society and thus population grows, resources per head become lesser
and lesser. Hence, there is a great and in fact immediate attention must be paid to the
better and effective use of earthy resources. It will be even better if we stop using the
resources, which is of course impractical. In addition to this, energy is used at every stage
of material cycle, which it self is a resource. That means that energy supply is limited,
and energy need to be used conservatively, and effectively. Moreover, during extraction
and production of materials / products, interaction with environment is usually
detrimental to the society in many ways. It includes water pollution, air pollution, and
landscape spoilage. Thus great care needs to be taken during extraction, manufacturing,
and use of a material / product. At end of its service, product may be recycled instead of
disposal. There are many benefits from recycling those include: to conserve raw material
resources and to conserve energy required to extract and refine. Recycling also avoids
environmental pollution. Hence, final products are needed to be designed and
manufactured such that they are environmentally friendly, and easy to recycle with
minimum usage of raw materials and energy. In case of disposal into environment,
products need to be bio-degradable. This shows that material life cycle involves
interactions and exchanges among materials, energy and the environment including the
society.

There is a need for better solutions to the environmental problems. Thus there is
additional cost to the product if it is required to be produced under conditions such that
detrimental environmental effects are kept to minimum i.e. green product cost more than
usual products. Industrial approach to assess the environmental performance of products
is termed as life cycle analysis / assessment (LCA). LCA is explained in detail in forth
coming sections.

18.3 Recycling issues

As stated in earlier section, recycling of a product is important for many reasons.

Recycling and disposal plays an important role in material selection and design of
modern products. Recycling can be defined as unflawed entry of material into its cycle
after completion of life as a product for infinite number of times. Environmentally
friendly means product after disposal into the environment, must interact with it and get
deteriorated so that material returns to the same state from where it is extracted and put
into life cycle. For ease of disposal and recycling, product design should ease the
dismantling of system. Different engineering materials are recycled and/or bio-
degradable up to different extents.

Metals and alloys tend to get corroded up to some extent i.e. bio-degradable. However,
some of them are toxic. On the other hand, most metals and alloys are recyclable.
Ceramics / glasses are, however, are hardly recycled. It is because their raw materials are
inexpensive, and recycling process is time consuming and expensive. Plastics and rubber
are highly popular among public because of their inertness to the environment and
moisture. This same inertness makes them hard to bio-degrade. Thus, plastics are mostly
recycled, and just disposed through land-fills. Thermo-plastic polymers are easily
recycled up on heating to higher temperatures. On the other hand recycling of thermo-set
plastics is much more difficult. Hence these are usually disposed. Thus, there is a trend to
use alternative materials which are recyclable. Ex.: thermo-plastic elastomers in place of
traditions rubber.

18.4 Life cycle analysis (LCA) and its use in design

Materials extracted from the earth are needed to be taken care of for one prime reason - to
conserve material and energy. However, materials, energy and environment effects are
interrelated i.e. each one has impact on another either directly / indirectly. Industrial
approach to assess the environmental performance of products is termed as life cycle
analysis / assessment (LCA). In other words, LCA is the assessment of environmental
impact of a product across its life cycle. The complex interaction between a product and
the environment is dealt with in the Life Cycle Assessment (LCA) method. It is also
known Ecobalance. LCA systematically describes and assesses all flows to and from
nature, from a cradle to grave perspective.

Environmental life cycle of a product consists of all stages from raw material extraction
to its waste management. In environmental life cycle assessment, natural resource use
and pollutant emission are described in quantitative terms. One important reason for
undertaking an LCA study is that there are growing concerns about a variety of
environmental issues as expressed by public opinion, political bodies, and industry.
Environmental concern may be related to the long-term resource base of human societies
or may be more health related or it may be a concern for the natural environment as such.
The strength of LCA is that it studies a whole product system. Since a whole life cycle is
studied, LCA is not site specific. Thus, environmental impact cannot be modeled at a
very detailed level. Economical and social aspects are not included in LCA other than
when used as a basis for comparison. Risk is another aspect not dealt with in LCA.

LCA is a technique for assessing the environmental aspects and potential impacts
associated with a product by

- complaining an inventory of relevant inputs and outputs of a product system;

- evaluating the potential environmental impacts associated with those inputs and
outputs;
 - interpreting the results of the inventory analysis and impact assessment phases in
relation to the objectives of the study.

LCA is not only product-orientated; it is also quantitative and thus seemingly objective.
Thus, it was no longer necessary to reply on simple rules of thumb. With LCA came the
notion that it was possible to possible to quantitatively compare alternatives in order to
identify the environmentally preferred option. Moreover, it could deal with
environmental issues in a structured way and it could handle several environmental issues
at a time. Quantitative description of the environmental impacts of emissions and
resource using the three categories resources use, human health, and ecological
consequences is not an uncomplicated affair. Environmental problems are complex and
still not fully understood. As a consequence, different attempts at developing models to
describe environmental impacts have led to different results.

LCA can be used in many different ways in product development. Life cycle assessment
is a technique for assessing the environmental aspects associated with a product over its
life cycle. The most important applications are: (1) Analysis of the contribution of the life
cycle stages to the overall environmental load, usually with the aim to prioritise
improvements on products or processes, (2) Comparison between products for internal or
internal communications.

Product design and development has been seen as the principal area of application of
LCA since the early days of LCA. The focal point of LCA, the product, coincides with
that of the product design and development process. This explains in part why LCA has
been brought forward as a tool for environmental adoption of product designs. The other
part is that product development is seen as a decisive activity for achieving sustainability
in industrial society most of the environmental attributes are determined and built into
the product during the design stage when materials are selected and constructions
designed.

With respect to product design, there is a need to understand how a product impacts on
the environment. To develop truly sustainable products, it must be possible to assess
which design solution is environmentally preferable. LCA tools can help in this difficult
area of eco-design. Ten simple guide lines for eco-design are:

1. Do not design products, but life cycles;

2. Natural materials are not always better;
3. Energy consumption often underestimated;
4. Increase product life time;
5. Do not design products, but services;
6. Use a minimum of material;
7. Use recycled materials;
8. Make sure product is recyclable;
9. Ask stupid questions i.e. be suspicious about the established result;
10. Become member O2: an international network for sustainable design.
As the whole product lifecycle should be regarded at once, representatives from design,
production, marketing, purchasing, and project management should work together on the
eco-design of a new product as they have together the best chance to predict the holistic
effects of changes of the product and their environmental impact. Environmental aspects
which ought to be analyzed for every stage of the lifecycle can be seen as:

1. Consumptions of resources (energy, materials, water or ground)

2. Emissions to air, water or the ground
3. Waste and waste products
4. Miscellaneous (noise and vibration)

Having made up a list on which phase of the lifecycle has which particular environmental
aspect, these aspects are evaluated according to their environmental impact on the basis
of a number parameters such as extend of environmental impact potential for
improvement or potential of change. The evaluations can also take marketing and
marketing issues into account. According to this ranking the recommended changes are
carried out and are reviewed after a certain time. It must be done regularly to keep up
with annual review of environmental legislation by national and international watchdogs.